Sound: Enabled, 44100KhZ Stereo, YM2612 High Quality enabled
Categories
Does not use warps or passwords
Abuses programming errors in the game
Manipulates luck
Takes no damage
Plays on hardest difficulty
Ignores time lost to bonus effects
About the Game
Rocket Knight Adventures is one of the better platformers on Sega Genesis. In it, you control Sparkster, an opossum with a sword and a jetpack, in his quest to defeat the evil pig empire.
About the Movie
This movie uses the US version of Rocket Knight Adventures.
This movie makes extensive use of a few programming glitches to maintain full speed as much as possible. This movie manipulates luck to prevent bosses from using their longer attacks. This movie takes no damage. This movie ignores time loss due to bonus effects. This movie plays on the hardest difficulty.
This movie is on a different ROM version from the previous run due to the following reasons:
Previous movie was on a PAL version, but was not PAL optimized, so is slower and less entertaining
US Version has a more difficult Hard Mode.
The US Version is the most likely version to be familiar to viewers.
Deep Loner began an improvement run on the US version, and I wanted to be able to make direct comparisons.
This movie aims for fastest time.
Suggested Screenshot
Improvement Explanations
Velocity control
Sparkster cannot maintain maximum speed when on the ground, instead, if you hold right, his speed will alternate between 3px/frame and 2.75px/frame
In the air, Sparkster accelerates at the same rate as on, ground and his maximum speed when holding right is the higher of 3px/frame and the speed you had when you jumped.
Sparkster is able to jump immediately at the start of a new room or new level if the jump button is pressed at the right frame beforehand.
If Sparkster starts a horizontal rocket boost in mid-air, half of his speed will be lost immediately when it ends.
Half of Sparkster's speed will always be lost immediately when diagonal boosts end.
Better optimized diagonal boosts used in wall climbing, and navigating bumpy terrain.
Better optimized use of horizontal boosts.
Made use of jumping before ledges to save time on long falls.
Optimized swimming for least wasted movement, and best possible boost locations.
In normal levels, directional controls are disabled during sword swipes
Combat optimizations
Hitbox display allows better optimization of timing and placement of sword-swings
The damage box on a sword swipes moves from top-rear of the player to mid-low forward of the player -- in order to connect with a single target for the full length of the swipe, you or the target must be moving.
Sword swipes last 8 frames, and do 4 damage each frame to enemies within the area.
The crescent slash produced by a sword swing causes 12 damage when it impacts a target, and can damage multiple targets if it connects on the same frame
Orthogonal (up/down/left/right) boost charges cause 32 damage upon impact, and initiate a circular boost for the remainder of the boost time.
Diagonal (up-left/up-right/down-left/down-right) boost charges cause 16 damage upon impact, and initiate a circular boost for the remainder of the boost time
Circular (no-direction held) boost charges cause 1 damage per frame.
The jetpack can be charged for a charge boost while slashing continuously by releasing the attack button only for one frame between swipes.
With one exception, enemies that flash cause lag each time they take damage
Enemies with HP will continue to take damage, even while flashing
Enemies that are Hit-based will not take damage while flashing, but will still cause lag when hit.
Unlike the prior movie, I do not attack any lag-causing before they become vulnerable
Luck improvements
Manipulated the starting position of the Tank Turret to a most optimal one.
Manipulated Axle Gear And His Giant Pig Mech to walk backwards as little as possible
Manipulated Pig Heads in stage 6 to always drop a row of apples
Manipulated Axle Gear to use only the attack pattern that causes him to be vulnerable.
Level specific notes
Stage 1-1
This level was really amenable to bunny hopping - only one obstacle that can't be jumped over, and no ceilings limiting when and how you can jump to do so.
The trucks were kind of a pain, after they pass by in the background, they always reappear in the foreground a specific number of pixels off screen, so hitting them as early as possible requires travelling the same way as them, and then immediately changing direction. I think I did a pretty good job optimizing their destruction, though. Each of the two trucks has 8 hit points more than they did in the prior published version, by the way.
You're not supposed to be able to jump over the wall after the trucks, but if you have enough speed at the top of the rise before it, you can just barely make it.
Keeping speed was a challenge in the area where the pigs parachute in, because I'm actually travelling faster than the crescent slice the sword shoots out. But it's important not to have to rocketboost too early on the hills after, because otherwise I'll hit the hill sideways while jumping and loose all my speed.
The downward sloping branches provide a nice speed boost.
Unfortunately I have to lose all my built up speed to deal with the four unskippable walkers. Each of these walkers has 8 more hitpoints than they did in the prior published version, by the way.
The walkers were also a challenge to optimize, as they appear in different places depending on the player's location, and the third and fourth will only appear after the defeat of the first two.
The four unskippable pigs are defeated much more quickly than in the prior version, by swinging the sword just before they become vulnerable, so the projectile passes through the near left one (which catches the tail end of the swipe) and hits the far left one. The sword swipe ends just as I have enough charge built up to boost and catch the two pigs on the right.
The tank is defeated as quickly as possible here. Each time it recieves damage on the first frame that its invulnerability wears off.
After the bridge goes out, the tank turret's x and y position depend on sparkster's position at the time it appears. I use rocket boosts to damage it while it's still offscreen, but I had to be very careful to hit it on the trailing edge, as I did not want to change its direction because the top right corner was the most desirable place for it to come on camera.
Stage 1-2
Slight timesaver by boosting downward as soon as I'm lined up with the rocket item.
The positions of the balloons can be manipulated by changing the timing of my sword swipes, but there's enough of a delay in between that they can appear at any point and I would still be able to kill them on the first frame.
These balloons will all appear with the same delay in between them no matter what, but the dragon won't appear until the last balloon is offscreen. The balloons travel much faster offscreen when they are crashing after being hit.
The ending sequence won't happen until its head is offscreen, this can be expedited by delaying the last hit for a couple of frames so that it explodes in a different pattern.
Stage 1-3
I wait a bit before moving right so that my boost ends before I reach the flames on the floor, so I don't die in a fire, and can clear the first bump in the floor.
I boost diagonally the first moment that doing so allows me to reach the higher floor
Only a small part of these flame bursts can damage sparkster, and sparkster is invincible during rocket boosts
By boosting at the proper frame, I am able to save a frame in reaching the doorway.
This is the first place where the "start by jumping" trick comes in handy. It allows me to reach a good height for rocket boosting even before the screen fades back in.
I fall a bit before starting the third rocket boost, so that the boost ends before I overshoot the door. I still begin the boost at the first possible frame.
The next room is not very noteworthy. a slight bit of hopping, and slightly better boosts than the prior version.
I save several frames in the last room before the boss by boosting left before falling down. The boost bounces off the walls and flings me towards the door at high speed, while not significantly impacting my rate of fall.
This boss is the first place I lose any gameplay frames to the prior version. I have to hit the boss three extra times compared to the previous run, and it has a significant invulnerability period between hits.
Stage 2-1
The "start by jumping" trick saves quite a bit of time, as it actually allows me to reach the mini boss before I hit the water.
I'm not sure if this is the optimal boss strategy, but I do maximize sword contact on the swipes.
I save a great deal of time in the swimming portion by diagonally boosting as high as possible.
This is faster than boosting horizontally, because I maintain a speed of 4 pixels per frame for much longer than it is possible to maintain the horizontal boost falling speed of 5 pixels per frame.
The diagonal boost strategy also allows me to reach the tree vine much sooner than the previous run does.
I try to maximize speed gain from the sloping vines.
The rotating platforms that bring me behind the waterfall are always in the same position when you reach them.
I touch the platform which switches me in front of the water fall because I need the next platform to take me behind when I land on it.
At some point during the fight with the water snake, I somehow trigger a bug which allows me to hit it from the other side of the waterfall. I don't know what causes this, but I do know it saves quite a bit of time, as I need to be in front of the waterfall after the boss is destroyed.
Stage 2-2
In NTSC versions of the game, this first room lags for the entire time I'm underwater, because there are so many moving spiked things to keep track of.
Boosting upward as soon as possible, after swimming past the downward column saves more time than it takes to recharge the pack to boost diagonally past the next upward column.
The second room is slightly tricky to time the boost so I don't hit any of the spikes balls that make up the ceiling
Stage 2-3
This stage is primarily an autoscroller, but it can be prolonged by not defeating the last pig in the first set as soon as possible.
When hopping between mine carts, momentum preservation techniques are important to make sure you get to the next fast cart as soon as possible.
The second set of pigs takes the same amount of time no matter what, so I chose to let the last four pigs survive until the game kills them, as it looks funny that way.
Although the prior version also attacked the train non-stop, I improve this by getting as close to it as possible, in order to maximize sword contact. This saves a large amount of time. Also, the left+right glitch allows me to swing the sword left, while I'm moving right. Which saves quite a bit of time as I have to be constantly moving a little to the right so the train doesn't hit me and kill me.
It is possible to do one or two damage to the hand during the first stage of the boss, but it doesn't carry over to the second phase, and even if it did, it would cause more lag than it's worth.
The second phase is defeated as soon as (hand 1 HP + hand 2 HP) < 0. Discovering this saved me quite a bit of time as I could just boost into Hand 1 to lower its hp more, rather than going back to Hand 2 to finish it off.
The third phase is a little different from the first. I have to be moving left constantly to keep from getting hit, but I can't move left and swing right unless I'm in the air, so I have to keep jumping. Unfortunately even the lowest jump I can make takes me to a height where I won't hit it with the crescent projectile if I swing. Fortunately, though, by hitting jump and attack at the same time, I can shoot the projectile without having to swipe, which lets me get an extra 12 damage in, quickly.
Stage 3-1
I start by moving left so that I'll be able to clear the first bump when I jump after the boost ends. This lets me carry high speed quite far into the level.
The lava won't hurt me while I'm boosting, and I can jump the first frame I finish boosting to take no damage. This helps save time as I don't have to wait for the lava to rise and fall to get through a section with low platforms. (Note: This is not true of the fires in stage 1-3, the lava in later levels, or the large spikes in stage 5-6)
I did my best at carrying speed through the platforms hidden behind crystals, but I have to boost in midair on my way down in order to get past low platforms during a high lava tide faster
Stage 3-2
I make some much better diagonal and horizontal boosts compared to the previous version, so I never have to stop moving right.
The crab boss' HP is reduced to 1/4 of its current value when the large claw is defeated. The HP is reduced to 1/2 of its current value when the small claw is defeated. Its HP begins at 50. After defeating the large claw, it becomes 12. After defeating both claws it becomes 6. But it is faster to go through 6 invulnerability periods than to destroy both claws.
Also, the prior run wasted time damaging the boss before destroying either claw, and at least two damage must be done for the total HP after a claw is destroyed to be lowered
Stage 3-3
This is basically an autoscroller, but the game lags by 8 frames each time a bat is hit. So I don't hit any of the bats.
Stage 3-4
NTSC versions of the game lag by increasingly high amounts during the part of the fish boss where it sucks in platforms. PAL versions do not.
My strategy of attacking the piloting pig is slightly more optimal than the prior run's. I suspect that it is not totally optimal, but brute forcing it isn't viable, and I suspect that properly figuring out the most optimal strategy involves calculus that I haven't learned yet.
Stage 4-1
Because the only way to damage Captain Fleagle is to bounce his own cannon-balls back at him, and he stops throwing them for a moment whenever he is hit, I have attempted to minimize the time between the cannonballs impacting him on each cycle.
I also have made sure that no cannonball he throws at me does not hit him.
Stage 4-2
The first part of this stage was quite frustrating to improve, as the wind makes it impossible to keep speed from a rocket boost, but I saved a little bit of time by boosting through Captain Fleagle in such a way that I land on the nose of the battle ship afterward
Going back to the left underneath was much easier to improve, as I could make use of momentum preservation, and came up with a much faster strategy for the mini-boss
Stage 4-3
I preserve speed here as much as possible, but the forcefields are confounding. It is possible to manipulate them, but I am not sure what the mechanism for determining when they turn off is. There is more than likely a possible improvement here.
Stage 4-4
Not a whole lot to say. I attack the giant slinky pig, and Captain Fleagle much more optimally than the previous version.
Stage 5-1
The missiles appear only after the last pig is defeated (or otherwise offscreen), but as in stage 1-2, the pigs appear at set intervals from each other. Also as in stage 1-2 their positions can be manipulated, but also as in Stage 1-2, the manipulation is unnecessary.
It's the boss from gradius!! I lose a lot of time here, though, because attacking it optimally requires taking damage, which I can't do.
Stage 5-2
Momentum preservation wins the day once more.
Momentum preservation, but vertically this time.
Stage 5-3
Better timing of diagonal boosts allows me to get through the bumps and dips faster
Stage 5-4
I take out the supporting pillar a couple frames sooner
I start the diagonal boost at the first frame where I'll actually be let in between the ceiling and the top floor from it.
Stage 5-5
I stop the platform's downward motion at the highest point where ducking will let me pass under the wall.
I jump at the first point where the wall ejects me rightward, but not down. This saves 44 frames.
A few more frames are saved by boosting sooner, and jumping from a higher platform
Stage 5-6
I start the diagonal boost sooner, and carry momentum to the exit by hopping.
Stage 5-7
I get to the giant robot boxing match faster, but it doesn't matter because I have to make 3 more hits on the Axle Gear's giant pig robot, and there's a significant delay between hits being possible to land
I also cannot take the damage necessary to land the final hit at the first possible frame, so I have to get in to make him start swinging, dodge out to avoid being hit, and then dodge back in so that he's within punching distance.
Stage 6
Wave 1: I'm able to defeat each set of 5 pig heads in a single swipe, by killing the master head before it generates the rest. I also collect the apple as soon as possible so the next set appears sooner.
Wave 2: I defeat this set of 8 pigheads quickly by making sure each swipe kills two. I then collect the apples as quickly as possible so that wave 3 will appear.
Wave 3: I am not sure if the number of asteroids onscreen influences anything, but I made it a minor goal to hit every asteroid in this wave at least once.
Wave 3: What does matter, though, is killing the false asteroids as soon as possible, as the next one won't appear until a certain time after the current one is defeated. Similarly, the next wave only appears once the last false asteroid is blown up.
Wave 4: You can't see it, but this wave consists of a bunch of pig heads moving in a circle. Only one needs to be killed for the wave to end, but killing the wrong head (or the right head at the wrong time?) causes a repeat of Wave 2 before the next wave comes, whereas killing the right head causes 6 apples to appear.
Wave 4: The next wave won't appear until all the apples are cleared. I collect all 6 apples before they reach the center of the screen, the prior run was only able to collect 4, and had to wait for the rest to move off the screen of their own accord
Wave 5: This weird guy can't be damages while he's holding his shield out to the left of him. But once he's dropped it I make quick work of him by maximizing each swing's connect time. The prior run didn't defeat him until in his third cycle, despite him having less HP in that version.
Wave 6: This is a repeat of wave 4.
Wave 7: Most of the pig astronauts don't even make it onto the screen before dying.
Wave 7: Axle Gear isn't vulnerable until after he's left the screen after firing the huge energe beam. The previous run just caused a whole bunch of avoidable lag by attacking him while he's firing the beam.
Wave 7: When Axle Gear does become vulnerable, I reduce his HP by 84 before he even makes it on screen. Unfortunately this delays killing the last two pig astronauts a bit, and there may be a possible improvement along those lines.
Wave 8: This is a repeat of wave 4 again.
Boss 1: This boss is composed of a whole bunch of separate objects with at least 120 HP each. All four turrets must be destroyed before the hatch becomes vulnerable. The hatch must be destroyed before the nose becomes vulnerable.
Boss 1: The nose's underbelly is the only multihit-enemy in the game that does not cause lag when it takes damage.
Boss 1: I have to hit the front of the nose once so that I can get in and hit the separation point without dying. This may look like a wasted swing, but if not for that, I would die upon entering that hitbox.
Boss 2: This is the most annoying boss in the entire game. He has a 19.5 second cycle between becoming vulnerable, and I have to get one more hit in on him than the prior run did.
Stage 7-1
Arrive at Emperor Devligus Devotindos earlier than the prior run, thanks to momentum preservation.
Emperor Devotindos is only vulnerable to rocket boosts, but I still manage a faster battle due to using straight boosts, rather than diagonal ones.
Stage 7-2
I get through this area full of Emperor Devotindos prototypes faster by better momentum control.
Stage 7-3
I enjoy totally outclassing Axle Gear
If Sparkster is facing Axle Gear when he stops flashing from taking damage, he'll perform a different attack which significantly prolongs his invulnerability. In order to hit him the first possible frame each time, I must jump, shoot the crescent projectile on the way down, and change facing direction for 2 frames as soon as I land.
Axle Gear also backs up by 1 or 2 pixels each time he is hit, I have to creep forward or eventually my crescent projectile will only be able tohit his crescent projectile, and not him as well.
I defeat Axle Gear's second stage much more quickly through better use of boosting, maximizing sword contact, and using two tricks that allow me to fire a projectile without making a full swipe (when I jump, and when I land).
Stage 7-4
I lose time here because I have wait to hit one of the emperor prototypes and boost through it, rather than taking damage.
Stage 7-5
I saved a few frames by timing my boosts and jumps better.
Final Boss
I lose time here because each of this boss's stages has 1 extra hit point, there's no quicker vulnerability cycle to manipulate, and the prior run never allowed the boss to be vulnerable for more than 2 frames before hitting it.
I like to think I was at least more entertaining during the wait time than the prior run, though.
Escape Pod
Saved a couple frames over the prior run by momentum preservation after boost
Reentry
I can't take damage, and I can't attack the Pig Star core in any way, so I tried to mock it by mimicking it's movements. It didn't work as well as I hoped, but better than just dodging for a couple minutes with minimum movement.
Comparison Table
Level
Start
End
Old Start
Old End
Input Length
Old Input Length
Lag Length
Old Lag Length
Level Savings
Level Loss
Total Savings
Total Loss
Menu
0
589
0
574
339
369
250
205
30
45
30
45
Stage 1 Cutscene
589
793
574
767
102
104
102
89
2
33
32
78
Stage 1-1
793
4347
767
5415
3246
4354
308
279
1125
*45
1157
123
Stage 1-2
4347
6776
5415
7869
2279
2313
150
141
34
9
1191
132
Stage 1-3
6776
11493
7869
12756
4393
4707
324
180
+458
+321
1963
453
Score Tally 1
11493
12532
12757
13783
962
962
77
64
0
13
1963
466
Stage 2 Cutscene
12532
12688
13783
13929
51
146
105
93
95
12
2058
478
Stage 2-1
12688
15113
13929
17249
2167
3072
258
248
*921
*26
2979
504
Stage 2-2
15113
16007
17249
18300
538
960
356
91
422
265
3401
769
Stage 2-3
16007
24753
18300
27315
8091
8320
655
695
*149
*40
3550
809
Score Tally 2
24753
25743
27315
28304
962
962
28
27
0
1
3550
810
Stage 3 Cutscene
25743
25823
28304
28375
17
18
63
53
1
10
3551
820
Stage 3-1
25823
26487
28375
29186
589
749
75
62
160
13
3711
833
Stage 3-2
26487
29114
29186
33007
2178
3264
449
557
1086
-108
4797
725
Stage 3-3
29114
31473
33007
35393
2269
2295
90
91
26
-1
4823
724
Stage 3-4
31473
33800
35393
37619
2055
2145
272
81
*98
*199
4511
923
Score Tally 3
33800
34816
37619
38630
963
965
53
46
2
9
4513
932
Stage 4 Cutscene
34816
35181
38630
38970
184
187
181
153
3
28
4516
960
Stage 4-1
35181
36642
38970
40463
1374
1418
87
75
44
12
4560
972
Stage 4-2
36642
39061
40463
43375
2225
2766
194
146
541
56
5101
1028
Stage 4-3
39061
39698
43375
44029
562
590
75
64
*36
*19
5137
1064
Stage 4-4
39698
42078
44029
46674
2231
2480
149
165
257
*-8
5394
1056
Score Tally 4
42078
43121
46674
47708
963
966
80
68
3
12
5397
1068
Stage 5 Cutscene
43121
43378
47708
47953
153
154
104
91
1
13
5398
1081
Stage 5-1
43378
49138
47953
53500
5211
4952
549
395
*@-218
*195
5180
1276
Stage 5-2
49138
50033
53500
54516
802
939
93
77
137
16
5317
1292
Stage 5-3
50033
50685
54516
55178
575
599
77
63
24
14
5341
1306
Stage 5-4
50685
52960
55178
57450
2198
2207
77
65
9
12
5350
1318
Stage 5-5
52960
53629
57450
58163
590
647
79
66
57
13
5407
1331
Stage 5-6
53629
57224
58163
61800
3519
3572
76
65
53
11
5460
1342
Stage 5-7
57224
60390
61800
64924
3034
3019
132
105
+125
+167
5585
1509
Score Tally 5
60390
61427
64924
65797
964
811
73
62
-153
11
5432
1520
Stage 6 Cutscene
61427
61703
65797
66056
171
154
105
87
17
18
5449
1538
Stage 6 Wave 1
61703
62137
66056
66614
430
555
4
3
125
1
5574
1539
Stage 6 Wave 2
62137
62370
66614
66954
233
340
0
0
107
0
5681
1539
Stage 6 Wave 3
62370
63713
66954
68472
1343
1518
0
0
175
0
5856
1539
Stage 6 Wave 4
63713
63832
68472
68693
119
221
0
0
102
0
5958
1539
Stage 6 Wave 5
63832
64462
68693
69829
630
951
78
185
*337
*-91
6295
1448
Stage 6 Wave 6
64462
64621
69829
70072
159
243
0
0
84
0
6379
1448
Stage 6 Wave 7
64621
65910
70072
71743
1062
1125
227
546
63
-319
6358
1129
Stage 6 Wave 8
65910
66221
71743
72155
305
408
6
4
103
2
6461
1131
Stage 6 Boss 1
66221
70627
72155
76740
3896
4117
510
468
*273
*-10
6734
1121
Stage 6 Boss 2
70627
76547
76740
81530
5889
4763
31
27
+0
+1130
6734
2251
Score Tally 6
76547
77578
81530
82399
965
813
65
56
-153
9
6581
2260
Stage 7 Cutscene
77578
77841
82399
82648
165
167
98
82
2
16
6583
2276
Stage 7-1
77841
79847
82648
84859
1704
1897
302
314
*314
*109
6897
2385
Stage 7-2
79847
80442
84859
85540
522
525
73
56
3
17
6900
2402
Stage 7-3
80442
83572
85540
90846
2526
4299
604
1007
+1871
+-305
8771
2097
Stage 7-4
83572
84259
90846
91453
595
535
92
72
@-60
20
8711
2117
Stage 7-5
84259
84638
91453
91825
307
311
72
61
4
11
8715
2128
Final Boss
84638
93284
91825
99159
8527
7263
119
71
+0
+1312
8715
3440
Escape Hatch
93284
93971
99159
99839
618
620
69
60
2
9
8717
3449
Reentry Cutscene
93971
95148
99839
101003
1092
1092
85
72
0
13
8717
3462
Reentry
95148
99408
101003
105251
4181
5182
79
66
1001
13
9718
3475
Total (Input)
0
98288
0
103005
90079
95115
8209
7890
5036
319
(N/A)
(N/A)
Total (Gameplay)
0
99408
0
105251
91131
97298
8277
7953
6167
324
(N/A)
(N/A)
Note: "Level Loss" and "Total Loss" are not simply differences in lag frame counts, I wanted to make the "loss" counts a sum of total frame loss due to version differences
"*":This level contained non-skippable enemies with higher HP in my version, I totaled the difference, and divided by average damage per frame, then added that number both to level savings, and level loss
"+":This level contained a "1-hit per cycle" boss with higher HP in my version, I took the number of frames per cycle multiplied by the number of extra cycles I had to wait on, and added that number both to level savings, and level loss
"@":On this level I lost time due to inability to take damage, but because of various uncertanties, I was unable to calculate how much time I lost due to that Final Frame Savings: 5839 Input Frame Savings: 4717 (input is ended later than the prior movie because I need to avoid damage longer during the reentry sequence)
Possible Improvements
Changing difficulty -- I learned too late that the only difference hard mode makes is in terms of Boss HP, and the amount of damage Sparkster takes from enemy attacks.
Changing versions -- The Japanese version of hard mode allows the player to take some hits without dying, and has decreased HP on some enemies and most bosses.
Lag management -- the game lags quite often, and I may not have found the most optimal way to deal with it.
Better Boss fights -- there are a couple of boss fights, especially the lava fish, that I'm reasonably certain I didn't find the optimal attack strategy, but couldn't find anything better.
Thanks to
CleverDan, for making an old run I could compare to.
Deep Loner, for demonstrating the real TAS potential of this game, and explaining the many new tricks he found.
Nitsuja, for adding autohold into Gens. This really saved my fingers when I had to hold attack and a direction for long periods.
Nitsuja again, for helping show me how to make interesting tools for Gens, like Ram search/watch, camhacks, velocity displays, Layer toggling, and more...
Hawq, for developing gens trace, which allowed me to disassemble portions of the game when I needed to figure out the mechanism behind something.
The Sonic hacking community as a whole, for all their awesome articles on Object Status Tables, which helped immensely when I decided to make a Hitbox and HP display tool for this game
Moozooh, Adelikat, Cardboard, JXQ, Emu, Titus Kwok, mmbossman, ShadowJacky, Raiscan, LagDotCom, and especially Derakon (who registered on these forums just to post that my WIP looked good) for their support in the creation of this run.
NesVideoAgent: Hi! I am a robot. I took a few screenshots
of this movie and placed them here.
I'm not sure I got the right ROM though. Well, here goes! Feel free to clean up the list.
adelikat: Accepting for publication as an improvement to the published movie.
Obvious yes vote.
I usually watch improvements simultaneously with a second emulator. While it is often useless, as improvements are barely noticeable, comparison was impossible this time, as you defeated the second boss before the first one was destroyed in the old movie :)
Great improvement and very enjoyable watch, good job!
I already stated my worries in IRC, that inexperienced viewers might be pampered by the fast action in the first levels compared to the autoscrollers later on (which upthorn handled very well) and so suggest to also watch the old movie to not miss the great variety of well executed improvements.
I liked it, gets a yes vote from me. Thought it would be a little faster if you lost some life but I suppose i can see why you did that. Certainly would be imposible to do on console. o-o
Just in case I wasn't clear enough in the submission text, in the US version, all enemies and bosses kill you from full health in one hit. Taking damage to save time would be faster, but is impossible on this version and difficulty level.
How fleeting are all human passions compared with the massive continuity of ducks.
Emulator Coder, Site Developer, Site Owner, Expert player
(3576)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
Magnificent improvement. It is a shame this game has so many auto-scrollers. The non-scrolling levels are amazing. The scrolling levels are pretty good. I like how you handled the scrollers though.
Obvious yes vote.
Joined: 11/18/2006
Posts: 2426
Location: Back where I belong
Excellent movie, I was bummed when I heard that Deep Loner had left this after barely starting it, so I'm glad you picked it up and ran with it. Great improvement, and entertaining throughout.
None of these improvements apply to the SNES spinoff, either.
The sequel to this game, is "Sparkster: Rocket Knight Adventures 2", for Sega Genesis. Sparkster for SNES is a spinoff which takes place on a different planet, has different characters, et cetera; but actually plays more like the original than the actual sequel.
How fleeting are all human passions compared with the massive continuity of ducks.
Congrats upthorn, you are very very strong. LOL Yes vote.
2 comments: In the first swimming level, at about 4:15 ish, you do one zip up, and then wait. Could the zip have come later to get over the middle piece?
Second: was zipping disabled for the flashy things with the moving lift? You stay on the platform the entire time, which is out of place in this ADHD movie.
Also, I thought I remember this movie ending at 26, not 27 min.
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
2 comments: In the first swimming level, at about 4:15 ish, you do one zip up, and then wait. Could the zip have come later to get over the middle piece?
First, for the sake of precise terminology, let us reserve the word "zip" to refer to abuse of a common wall-ejection mechanism (wherein if you collide with a wall, the game begins moving you at high-speeds, backwards relative to the direction you hold down)
Now, to answer your question, I could have delayed the rocket boost by a few seconds so I could swim high enough to get a diagonal boost over the middle wall, but this would have taken a few seconds longer. The boost currently ends about a half-second before reaching a height where I could boost diagonally to get over the wall, so I only have to wait around 30 frames to do the next boost, compared to the extra 3 seconds it would have taken to swim to that height.
DarkKobold wrote:
Second: was zipping disabled for the flashy things with the moving lift? You stay on the platform the entire time, which is out of place in this ADHD movie.
Rocket boosting is possible during those sequences, and I actually do currently rocket boost at the first possible opportunity where I can reach the exit by doing so. But for the first room, the exit is placed so high that it can't be reached by boosting, even from the peak of a maximum height jump, so I need to wait for the platform to get nearby.
Sidenote: There is actually a small wall zip in this room. I jump on the first frame that the wall ejects me to the right, instead of downwards, which saves 44 frames compared to waiting for the platform to reach that position.
For the second room with a moving platform, the floors ceilings, and walls are made of instant-death spikes. If that were not the case, it would be much faster simply to walk than to take the moving platform, but the instant death spikes make walking, and even boosting, impossible.
How fleeting are all human passions compared with the massive continuity of ducks.
Fast levels were yes. Autoscrollers were good, but too many of them makes them meh. Graphics and especially sound are quite good for the Genesis. Giant robot pig who thrusts his pelvis forward to attack = giant win. Boss who you fight by getting in his spare giant attack pig = win^2. Amount of bosses = ridiculous. It got a little monotonous at the end due to that, but overall the plusses smack the minuses, and I vote yes.
<Swordless> Go hug a tree, you vegetarian (I bet you really are one)
Thought it would be a little faster if you lost some life but I suppose i can see why you did that.
Just in case I wasn't clear enough in the submission text, in the US version, all enemies and bosses kill you from full health in one hit. Taking damage to save time would be faster, but is impossible on this version and difficulty level.
I may have overlooked it, I thought I read the submission text. Wow, then beating it on hard is definitely a worthy accomplishment!
This is one of my favourite genesis games. The speed at which you cut down those bosses is amazing.
I was constantly wondering "how the hell did he do that?". One of the better movies on this site imo.
In order to speed up the publication process, I want to mention, that by obsolescing the published movie, the last genesis entry in the Lowest technical rating will be deleted, proving apodictically for once and for all, that gens tasers are not only more beautiful and sexy, but also the better tasers ;)
Yay for the rocket mouse hero who moonwalks (sparkster jackson?) all along the run to save his princess...
I loved this game and the run is nice, I voted Yes !
Nice job...
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Hey, I was just reading the submission text and got confused. One of the massive improvements in the run came from preserving the boost speed from only touching the ground for one frame and bunny hopping through levels... is that already in there (and thus difficult to understand since I couldn't see it) or does it need adding in?