Tool-assisted game movies
When human skills are just not enough

Submission #6837: Winslinator's INTV Sharp Shot "maximum score" in 04:05.24

Console: Intellivision
Game name: Sharp Shot
Game version: any
ROM filename: Sharp Shot (1982) (Mattel).int
Branch: maximum score
Emulator: BizHawk 2.4
Movie length: 04:05.24
FrameCount: 14695
Re-record count: 22001
Author's real name:
Author's nickname: Winslinator
Submitter: Winslinator
Submitted at: 2020-08-07 23:39:30
Text last edited at: 2020-11-25 15:29:40
Text last edited by: feos
Download: Download (9803 bytes)
Status: decision: rejected
Submission instructions
Discuss this submission (also rating / voting)
List all submissions by this submitter
List pages on this site that refer to this submission
View submission text history
Back to the submission list
Author's comments and explanations:
Most up-to-date movie file: http://tasvideos.org/userfiles/info/65254876577894993

Sharp Shot is an odd yet simplistic game released in 1982 for the Intellivision. It features four minigames: Touchdown Passing, Space Gunner, Submarine, and Maze Shoot—each of which were likely old demos for NFL Football, Space Battle, Sea Battle, and AD&D: Cloudy Mountain, respectively.

The goal of each game is to shoot as many targets (or touchdowns) as you can in 60 seconds. Though it is a score-based game, time still comes to play in the TAS through luck manipulation at the start of each minigame and how quickly the game is switched after the timer runs out.

(Link to video)


Okay, so what exactly was the goal here?

There are four minigames, each of which can be selected at any time. To make this a complete submission, I do not switch minigames until the timer completes for each as this will effectively showcase all of the game's content.

Now, I aim for the maximum score in each minigame, but since the minigames are on a fixed timer, where does the speed challenge come from? Well, points can still be scored after the timer ends so it is preferrable to accumulate those last points right as the timer ticks to '0'. Additionally, some of the minigames (Submarine, most notably) have a delayed start for luck manipulation.

2 Players Only Space Gunner and Maze Shoot allow both players to control at the same time so I took full advantage of that during this TAS. I did not play Touchdown Passing and Submarine twice to give the second player a turn as that seemed redundant and would detract from the entertainment value of the run.

Touchdown Passing

The paths that the quarterback, receivers, and defenders run are all random. To score as many touchdowns as we can as quickly as possible, we need to manipulate RNG such that the quarterback and receivers quickly line up with each other, and the receivers don't run too far into the endzone as that will waste time. The player only controls when the football is thrown, and it always travels in a straight line.

It appears the minimum number of frames between touchdown passes is 220. I continually improved the longest segments until no faster variations were found. The final product has no duration between passes that exceeds 240 frames.

Space Gunner

Shoot as many spaceships as you can. There can be up to five ships on screen at once, and luck is a major consideration here. Where the ships fly in from is random and I generally manipulate such that no ships spawn from the upper left or fly down the middle, as these are the longest flight paths to the crosshairs. Both of these paths cross P1's crosshairs first, hence the heavy score bias towards P2. Ships only come from these directions in the final movie if manipulating otherwise was slower or not possible.

Explosion shrapnel is random and can blow up additional ships in the vicinity. I fire on the right frames to blow up as many additional ships as possible ouside the crosshairs. Only five ships can be on screen at once, so the sooner I can shoot every ship, the sooner more can spawn.


New ships can spawn on the left or right side of the maps and have varying speeds on the paths they take. New ships are ideally spawned in the direction the submarine is headed so they will meet up the quickest. There are four submarines, so the faster you shoot a sub, the faster you will be able to shoot the one that respawns.

I waited 21 frames before switching to Submarine and 24 before starting to start spawn four ships on the left. Yeah, luck manipulation in this minigame is extremely tough, as you may have guessed by the odd, unoptimal timing of many of my shots. Only three torpedoes can be fired from the sub at once, adding to the madness.

Maze Shoot

The methodology here consisted mostly of shooting whichever enemy or groups of enemies took the least amount of time to shoot. I frequently go out of my way to shoot groups of enemies that are close together for a precise double or even the rare triple shot! The player cannot control where the arrows shoot from; only when they fire.


Game Score
Touchdown Passing 112
Space Gunner 68+113
Submarine 71
Maze Shoot 79+69
Total 512
I move on to the next minigame as soon as possible after the max score is reached for each. In Submarine, the final score is only on screen for one frame so it is barely perceptible. Score is tracked at Main RAM addresses 60 and 61 for P1 and P2, respectively.

Suggested Publication Notes

Insert first two paragraphs in these submission notes here

Winslinator shows off some superhuman marksman skills in each minigame, attaining the highest-known total score of 512 points in record time.

Suggested Screenshot Frame 12677

feos: Judging...

feos: Replacing with 10-point improvement.

feos: Feedback was kinda mixed, several people liking the movie and several people disliking it. When I think of some spherical average user in vacuum, I can see why this movie may not be entertaining enough to get into Moons. I explained my personal take here. Not that I don't want this movie in Moons because I disliked it, I just think the feedback was too mixed to take sides. Which means we can't say that this movie is generally quite entertaining.

But there was some talk about whether this goal should be Vault eligible. As Moth explained, this movie can't be considered a speed record in its nature, because it doesn't aim to end input early. Instead it uses all the available time to gain the highest possible score. Without the max score goal, one could just wait or occasionally shoot some objects, and the game would still end at the same time. This movie is not a time attack, it's a score attack, and nothing more. While Vault's "focus is exclusively on collecting tool-assisted speedrun records".

We had some extra discussion among staff about whether we want Vault to include more categories or maybe just include timed games in some form, maybe only for full completion.

We agreed that we don't want timed sports to be added, because of complexity and variety of their rules, which gets especially problematic when we try to define full completion or max score for them. Some games include variety of barely relevant stats one could maximize. In some games the timer can be paused. Some of such movies may last for several days. And none of that brings anything valuable to the table if we remember that Vault rules try to be as straightforward as possible, and the records should be widely considered meaningful and legitimate. And defining straightforward and agreeable rules for this inherent complexity looks like too much work for too little benefit.

Defining fastest completion for timed games is also a weird concept. You can end input early, but you can't make the ending happen sooner by better gameplay, aside from marginal optimizations like lag reduction. It could be argued that due to minor differences in how soon you can stop the movie or make the game end, such a goal is too trivial to compete for. There's just not enough room for gameplay optimization which is meant to be unlimited in terms of resulting time.

But even if we ignore timed sports games and full completion for timed games, there's still a fundamental question about what's left: full/max completion for timed games and Vault eligibility of that.

Full completion has a problem that after you've collected all of your items (or whatever it is you're completing), you will have to simply wait out the game doing nothing useful. In some games you need to do some basic things to still avoid a game over, but in others it's even simpler, so it may be many minutes of nothing meaningful at all. Even if we end the movie right there, we still have to sit through the rest, be it minutes or hours. And of course in some cases the game is so simplistic that ending input early makes it trivial to speedrun complete.

The final option is only allowing movies like this one, which uses all the available time to score to the maximum. And once again we couldn't come up with strong and convincing reasons to change the nature of Vault, why it would be good for everyone, and why such a thing would fundamentally fit.

The reason Vault even exists is to provide room for speedrun records that Moons can't keep due to low entertainment value. Since the site always aimed to host entertaining superplay movies, most of them being speedruns and some of them being playarounds, technically impressive speedrun records that were boring were being rejected until 2012. When adding a tier for boring stuff, we aimed to limit it to the most simple and sensible goals ever existed: fastest completion and full completion. If something is not speed oriented, it should be entertaining enough for Moons. If we don't limit such things, Vault would have to include all sorts of esoteric goals, requiring a lot of extra work on all levels, while reducing the overall quality of our publications. There was a lot of discussion about allowing boring, non-speedrun movies into a new tier, Demonstration, but over the years we haven't even once agreed about the rules such a tier should have.

A few of us wish to expand Vault for timed games (including me), but that feeling isn't a good reason in itself, if it's not based on strong arguments against the current system, that would look justified and be supported by site admins and wide audience. Wasn't the case here.

As a result, rejecting.

Similar submissions (by title and categories where applicable):