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

Submission #5759: qflame's SNES Math Blaster Episode 1: In Search of Spot in 07:53.15

Console: Super NES
Game name: Math Blaster Episode 1: In Search of Spot
Game version: USA
ROM filename: Math Blaster - Episode One (USA).sfc
Emulator: BizHawk 2.2.1
Movie length: 07:53.15
FrameCount: 28436
Re-record count: 1854
Author's real name: Jacob Beckley
Author's nickname: qflame
Submitter: qflame
Submitted at: 2017-12-22 21:41:20
Text last edited at: 2018-01-29 21:45:10
Text last edited by: ThunderAxe31
Download: Download (11595 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:
This is a run of Math Blaster with the main goal being speed. Math Blaster has 3 sections: Trash, Caves, and Ship. In the menus you can choose between 3 and 5 lives. I add 1 life because I will be taking a lot of damage in trash.

A video of the run can be seen here:

(Link to video)


In trash it is better to hit a wrong answer first as it only takes an additional 19 frames to fire the second shot, but hitting a correct answer first adds 40 frames because your bonus points increase and they have to be tallied in the bonus stages. For this reason, I hit an incorrect answer first all 30 times. This is the reason for the additional life. In the bonus stage, I avoid all bonus points except at the end, where shooting the last asteroids can end the stage sooner.


The main feature of caves is the up-clip glitch which was discovered recently. Normally, you are not able to progress upwards unless the number on your character is between the minimum and maximum listed on the cave level. However, with this frame perfect and pixel perfect glitch you can avoid getting detected by the laser that usually shoots you down.


In ship I manipulate the RNG to only give me problems where the correct answer is in the 2 leftmost tubes by waiting a specific amount of frames. The RNG is entirely based on the frame number and there is no other known way to manipulate it. The second tube is only slower by 1 frame, so both are acceptable as it would take at least 1 frame to RNG manipulate again until the correct answer was in the first tube. The only other glitch used here is known as the "Owch clip" where there is a 1 frame window to get "owched" by a piece of trash flying by and still make it into the tube.

ThunderAxe31: Judging.

ThunderAxe31: Hello and welcome to TASVideos!

This run is not acceptable for Moons tier because the entertainment value is very low. Thus, Vault tier requirements will be applied.

The run is nicely optimized and beats all known records. However, the game played does not meet the Vault requirements, since it's an educational game. Specifically: this game mostly consists in doing math fast, and in fact most of the efforts showcased in this run are actually just manipulating and predicting the answers in order to solve the problems fast.

For this reason, I'm rejecting this submission. Better luck next time!

ThunderAxe31: In view of the arguments provided by feos in the forum thread of this submission, I start over the judging process.

ThunderAxe31: The argument brought by feos consisted in a different interpretation of the Vault rule for educational games. While I considered that rule to forbid any run made with educational games, he did instead consider it as actually forbidding games that don't feature TAS-worthy material. Since his argument was supposedly supported by the goal of TASVideos of developing superhuman gameplay, I decided to consider the possibility that my initial judgement was wrong.

I had a conversation with other staff members, including Nach, Mothrayas, and feos. I explained the reason for my judgement and I presented my evidence pointing out that Math Blaster can't be completed casually without solving math. In the end, everyone acknowledged that my method added a clear cut to the rule, whose text was updated accordingly. We needed to draw a clear borderline for evaluating if a given title is primarily an educational game or not. My idea was to use the concept of "casual play" as a yardstick for estimating how much determinant is the requirement to perform educational activities in order to play through the game.

On the other hand, we also agreed that relying on TAS merits for a given run could never give a definitive extimation, since that substantially consists of speculating about the TAS potential available for a given game. We can't really know in advance if such potential is actually present, and that would result in relying on chances, which we can't do for Vault rules. In fact, TAS potential can be there, but until one tries hard enough, we won't know about it. This doesn't allow for any reliable rule.

It must also be noted that while it's true that the goal of TASVideos is to develop and showcase superhuman gameplay, this is mainly done for the purpose of entertainment, which clearly doesn't apply for the Vault tier. And on the other hand, this movie has been proven by the audience to lack any TAS merits that make it entertaining to watch.

The purpose of the Vault tier is keep track of videogame records, and thus shouldn't be applied for pieces of software that can't be considered as actual games. For this reason, some kinds of titles are excluded from Vault tier, like educational games. Even if the updated rule that defines an educational game is quite lax, it's still very clear and definite, and it must be so in order to avoid impossible-to-solve cases; raising an exception here would generate a bad precedent.

This is indeed an unfortunate case because the run itself features good TASing material, as explained by feos in this post, and I also was aware of this from the start; but even then the run was not entertaining enough to be accepted for Moons. The best I can do is to note that a "maximum score" run could potentially be entertaining enough to be accepted. Lastly, I want to thank qflame for having submitted this run, because it did give the opportunity to test and refine the rule.

Reassuming: the rule didn't change, my judgment didn't change. Rejecting again for bad game choice in conjunction with low entertainment.

Similar submissions (by title and categories where applicable):