(Link to video)
Download encode here
You play as a submarine that is entirely unrelated to the UEO seaQuest DSV 4600. In normal gameplay, you get 10 * (1 + currentLevel) points for destroying a fish or an enemy sub, 50 * currentLevel points for each diver you rescue, and 10 * (1 + currentLevel) * your remaining oxygen when you rescue the divers. After reaching level 8, using the No-Despawn glitch, I collide with a sub for 3 minutes until I reach the maximum score of 999999.
This submission is in response to the feedback for #4291: morningpee's A2600 Seaquest "maximum score" in 00:06.02. Credit goes to VELHO for discovering the glitch.

Game Objectives

  • Emulator used: BizHawk 1.6.1
  • Fastest time to 999999
  • Manipulates luck
  • Ends input early
  • Uses a glitch

Luck Manipulation

The game uses a 217-index LCG to generate random numbers at address (02h). By paying attention to these values, you can control which side of the screen the diver comes from and avoid RNG values that result in no divers.
If divers are on-screen when you finish a level, the same number of divers will spawn immediately at the beginning of the next level, and at the same depths. you can wait a certain number of frames (unaffected by (02h) ) to control which side of the screen these divers spawn from.
By waiting, you can also control which sides of the screen fish will spawn from. This is also unaffected by the RNG at (02h).

Limitations on spawning divers

There are 4 separate depths that divers swim at. Generally, you must collect divers from all 4 depths before being able to spawn divers from the first depth again. For example, if you collect a diver at depth #2, you must collect divers from the three other depths before spawning a diver at depth #2 again. This detail limits the efficiency with which you can collect divers, and significantly impacts the routes I take.

No-Despawn Glitch

Normally, when you collide with a fish or enemy sub, the enemy despawns immediately, and your sub despawns 35 frames later. However, by colliding with an enemy in a special way, you are able to prevent the enemy from despawning, which keeps your sub's despawn timer from completing.
In this case, you get 10 + 10 * currentLevel points for every frame you are in contact with the enemy, for currentLevel <= 8. For example, if you perform the glitch on level 4, you receive 50 points for each such frame.

Optimizing the No-Despawn Glitch

There are two types of enemies: Fish and subs. Because each fish bobs up and down, there is no way to perform NDG (no-despawn glitch) on a fish so that you are in contact with it the entire time until 999999 points. It is possible with an enemy sub, however, as only its rudder moves.
The most significant problem is to decide which level to perform the glitch on. Consider these details:
  • Using a preliminary run as a basis, the average, optimized, luck-manipulated stage will take 747.5 frames to complete.
  • The first stage begins at frame 129.
  • Once you have performed NDG, you gain 10 + 10 * currentLevel points per frame until you reach 999999 points, for currentLevel <= 8. After level 8, you still get only 90 points per frame.
We can use an equation to model the number of frames it would take from each level to reach 999999, if x is the current level and y is the frame estimate. Note that we only need the integer part of x:
y = 130 + 747.5 * floor(x) + 999999 / (10 + 10 * floor(x) ), x <= 8
Minimum frames is at level 8. Note that the actual minimum of the function would be at x = 10 if it were not bounded, but that level 10 offers no advantage over level 8.
Using this method, it takes 4 minutes and 39 seconds to reach the maximum score of 999999. At this point, the enemy swims away (without ever despawning), and the kill screen is shown. In the kill screen, the "Activision" text at the bottom is replaced with "Copyright 1983", though you never actually die. This is the same behavior as when you reach 999999 by means of normal gameplay.
For comparison, see this [dead link removed] savestate, where 999999 is reached by normal means. After a few seconds, the seaQuest will collide with a fish, "Copyright 1983" will be shown, and your sub does not despawn.

Levels 1-7

These levels are completed as quickly as possible, using careful luck manipulation and route planning. At the beginning of level 5, a diver spawns at depth #1, so I don't need to kill the fish for diver #1 in level 4 for the carryover. In level 8, I don't need to collect any divers, so I don't kill any extra in level 7.
The barge that floats at the top of screen limits how many extra divers I have time to collect in level 5. I only had time for 1. It also limited how far left I could be at the end of level 7.

Level 8

In this level, my goal is to hit the correct point at the back of a right-to-left sub quickly. By returning to the surface with only 1 diver, I can cause the fish to swim faster, which will let me trigger NDG sooner. Frame 5870 is the soonest I can surface, or else the enemy sub that I use for NDG will spawn from the left side of the screen instead of the right, which I do not want. This is why the fish turn green before leaving the screen on level 8.

RAM Watch

RAM addresses are here.
To summarize, this max score TAS of Seaquest is comparable to Tetris because I am reaching 999999 as quickly as possible. It is also different from Virtual Pinball because there is a clear ending point here.

feos: Rejecting and publishing Accepting to Vault and publishing...

Noxxa
They/Them
Moderator, Expert player (4140)
Joined: 8/14/2009
Posts: 4083
Location: The Netherlands
feos wrote:
And the branch is put only for the sake of information. Probably it's not needed, but I don't see a problem with having it.
This is the problem I see with it:
I wrote:
it's still wrong to publish it with a non-vault branch name. At best, it's misleading about what branches are accepted in Vault and what not.
I wrote:
Those labels indeed describe the movie, but it's clear that their goal is still to get to the completion point in the fastest time. In this case, it isn't so clear, as "fastest maximum score" is a different goal than "fastest time", and is not a vaultable goal unlike the latter.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
Ok. So should it be vaulted back, with no branch?
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.
Noxxa
They/Them
Moderator, Expert player (4140)
Joined: 8/14/2009
Posts: 4083
Location: The Netherlands
I'm a bit on the fence about it - as I said, I find the "ending point" somewhat dubious, but it seems a fair amount of people agree it's a valid ending point. If enough people agree about the ending point, I suppose it can be put back to Vault without branch name.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Player (55)
Joined: 11/28/2013
Posts: 118
Mothrayas wrote:
I find it somewhat dubious. The little "copyright" blurb seems more like an easter egg if anything, and the game itself doesn't look like it ends.
Apart from the example of reaching 999999 (and then getting hit), the text "Copyright 1983" shows up if your sub dies and you have 0 lives left. Also, after you reach 999999 and get hit, your sub is uncontrollable, even though you have many lives remaining. I don't think the ending is an easter egg or a glitch.
Projects: Tetris DS Genesis Toys: Let the Toy Wars Begin
Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Based on the discussion, it appears to me this should be vaulted, with no specific branch name, and the description should mention that the game stops when score is maxed, and the sub is no longer controllable.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3599)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
Under goal choice in the Vault rules: "Must be clearly definable as having completed the game." If the best we can do for defining an ending is 999999 points, then this game fails to have a clearly definable goal. The intent of this rule is for exactly a scenario like this. If we find ourselves arguing over how to define an ending then it isn't vault-worthy. A consistent rule of the vault is that in the face of subjectivity, we demand entertainment value (therefore not vaultable, must be moon). Vault is minimally subjective. As such the only logical conclusion here is for this movie to be a moon. The controversy here is that people feel it has subpar entertainment value for current moon standards, but that's ok. That will happen from time to time and is less objectionable than a controversial vault movie. This movie was judged incorrectly. It should have been deemed not vaultable and judged by moon standards. Also, judging from the ratings, the movie isn't THAT subpar to the audience.
It's hard to look this good. My TAS projects
Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
adelikat wrote:
Under goal choice in the Vault rules: "Must be clearly definable as having completed the game."
The "clearly" I guess is the issue. It's pretty clear to feos. If it's clear, see what I said above. Otherwise, I agree with you.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
We don't define and ending at 999999. It's just the only place where the game acts like it ended (similarly to as you die). Kill screen or how that stuff is called. And I don't see anyone questioning the validity of this ending before it was published. "Must be clearly definable as having completed the game", as a rule, may appear to be hard for some Atari games, in such cases we once again can rely on the majority. Some people consider it clear, some don't. If most of them do, I'd agree with them.
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.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3599)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
And I don't see anyone questioning the validity of this ending before it was published.
Why would anyone do it pre-publication?? Your logic doesn't make sense. The run is a max score run, and judged and critiqued as such. Nobody was asked whether this run was a valid any% run, you made up the consensus here.
It's hard to look this good. My TAS projects
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
adelikat wrote:
The run is a max score run, and judged and critiqued as such. Nobody was asked whether this run was a valid any% run, you made up the consensus here.
The submission addresses this, and no one objected the ending point chosen the author.
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.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3599)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
This was advertised as a max score run.
It's hard to look this good. My TAS projects
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
adelikat wrote:
This was advertised as a max score run.
How does it disprove my previous post?
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.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3599)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
Nobody was considering whether or not this movie was an any% movie or not. I see no discussion regarding that until post-publication. Your views on what an audience is or isn't agreeing on is disturbing to me. I'm done arguing this point unless ONE other person supports you on this.
It's hard to look this good. My TAS projects
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
adelikat wrote:
I'm done arguing this point unless ONE other person supports you on this.
http://tasvideos.org/forum/viewtopic.php?p=377897#377897
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.
Fog
Experienced player (626)
Joined: 4/5/2014
Posts: 459
How about we just remove the branch name and put it back to vault? The game by design does not have an ending point. However, it does not allow the user to score more than 999999 points. It's essentially a kill screen of sorts because you're unable to get more points. Any TAS that tries to improve upon this must reach the same score because it's the only known "end-point" to this game.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3599)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
feos wrote:
adelikat wrote:
I'm done arguing this point unless ONE other person supports you on this.
http://tasvideos.org/forum/viewtopic.php?p=377897#377897
Ok, we are discussing too many things at once. I meant nobody is supporting your notion that pre-publicatoin there was a consensus. The vaultability discussion was post-publication. Nothing pre-publication is viable for determining a consensus. Post-publication I see 3 people that think this is not any% or at least unsure. And 2 people who think it is. That's hardly a consensus. This was a judging error in my opinion. A max score run doesn't go into the vault. If it was deemed an any% that's a potentially controversial decision that should have been discussed beforehand. What should be done after is leave it as it is currently, in moons. The more the idea of this being any% is discussed the more it is clear that this isn't a non-controversial non-subjective idea. And therefore, not ideal for the vault. Some games (especially atari 2600 and platforms of that era) just aren't vaultable because it is difficult to define game completion. In the face of that kind of subjectivity, the rules simply require a run to have entertainment value.
It's hard to look this good. My TAS projects
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
adelikat wrote:
Nothing pre-publication is viable for determining a consensus. Post-publication I see 3 people that think this is not any% or at least unsure. And 2 people who think it is.
Myself, jlun2, scahfy and Nach agree with the ending point's validity. You and Moth don't. Moth is also letting the people decide. I don't know where you got your numbers from. I can agree though, that it was a mistake to not raise the exact question about this run completing the any% goal before publication. It just seemed quite clear to me (it still is), and no one saw an issue.
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.
Player (55)
Joined: 11/28/2013
Posts: 118
adelikat wrote:
This was a judging error in my opinion. A max score run doesn't go into the vault. If it was deemed an any% that's a potentially controversial decision that should have been discussed beforehand. What should be done after is leave it as it is currently, in moons. The more the idea of this being any% is discussed the more it is clear that this isn't a non-controversial non-subjective idea. And therefore, not ideal for the vault.
The run is improvable. Since this branch seems unentertaining and unvaultable, I don't see how an improvement will be publishable if the current publication remains in moons. I think I agree with feos, for the sake of consistency within this one game.
Projects: Tetris DS Genesis Toys: Let the Toy Wars Begin
Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
feos wrote:
Myself, jlun2, scahfy and Nach agree with the ending point's validity.
I don't agree with the ending point validity. I described earlier that if the discussion agrees with the validity, how the run should be handled. scahfy: Looking at the entertainment ratings this has gotten, some of them were clearly moon quality ratings. An improvement which can significantly shorten the run length would also imply less to get bored by, which can only help your ratings. If you can throw in some close to the edge near misses which demonstrate you precisely predict oncoming attacks, a better run may even be worthy of moon tier proper.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3599)
Joined: 11/3/2004
Posts: 4739
Location: Tennessee
Also, we would never reject an improvement to a published movie. Even if the publication is dubiously published, we would certainly rather have the better movie.
It's hard to look this good. My TAS projects
Player (55)
Joined: 11/28/2013
Posts: 118
Okay
Projects: Tetris DS Genesis Toys: Let the Toy Wars Begin
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
Pitfall II acts exactly the same after you beat the game. The character stands still, and the Activision/Copyright thing scrolls. It proves that this game really ends at reaching the maximum score.
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.
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
It seems all the fight was in vain, because we need difficulty and new content maxed out for games without clear ending.
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.