- Emulator Used: BizHawk 2.10-RC2, but the movie syncs on 2.9.1
Game objectives
- Completes Pocket Monsters: Green as fast as possible without save data corruption
Version choice, or rather, why Green?
...because this is the version name exclusive to the original Japanese release of the first Pokemon game. This makes it clear that it is this version being discussed. Admittedly, Red might be technically slightly faster with Charmander's cry on the title screen being a few frames shorter than that of Bulbasaur and the player's first default name having 3 characters instead of 4, but those are negligible differences.
Additionally, v1.0 is chosen because the primary glitch used for the route is only available there.
The warp
This route is the fastest at the start of the game. Note that it's not possible to initiate a warp from Viridian under such circumstances so returning to Pallet is required.
Improvements
The actual game-ending glitch set up is identical to what's done in the previous movie for most part. Actual improvements are from the portion before that.
First, not only is it slower to change game options right after title screen (77 frames) instead of on the overworld inside the player's room (58 frames), it's actually entirely unnecessary for this route as the total time loss from animations (50 frames) is shorter than that: Scratch animation is only 10 frames slower than the screen-shaking effect when the animations are turned off, and the Pokeball animation when the player's mon is sent out only takes 10 frames.
Gain:
- 66 frames from Tackle animation over Scratch animation (22 per use)
- 41 frames between moving to pick starter and the start of rival battle
- 4 frame from cry on the second battle (maybe another 4 frames at the end of the rival battle during fainting too)
Loss:
- 9 frames from cries during the intro of rival battle
- 72 frames from one more Growl miss
- 15 frames from one more crit
This means choosing Charmander is technically faster than choosing Bulbasaur by around 20 frames. Such advantage can easily be lost from manipulation needed for the additional Growl 1/256 miss and max range crit over regular max range, but I decided to go with this theoretical faster route with each of the 6 move manipulations done by Basic Bot of BizHawk.
Route
- Nothing is nicknamed as no name appears for enough time to warrant it. The player gets the second default name as it's 1 character shorter than the first one.
- Potion is taken out of the player's PC as the first item required to perform the Select glitch.
- Charmander is picked as the starter with 18 HP. To quote CasualPokePlayer:
"The Rival fight is purposefully lost, as it does not need to be won and it is fastest to lose. This is done in 3 turns, each comprising of a 1/256 miss with Growl, and a 1/39 crit with Tackle. A 1/39 Tackle crit does 6 damage to Charmander, so having 18 HP allows for a 3 turn fight. Who doesn't love a 1/1178114537369 Rival fight."
- Parcel is acquired in Viridian Mart as the second item.
- A Pidgey is encountered on the way back from Viridian to Pallet to do the Select glitch. Pidgey is preferred over Rattata for a shorter cry (6 frames). It is also deliberately made to happen at a tile such that the number of steps it takes to reach the last grass tile in Pallet (second tile on the Pallet map) is a multiple of 4. This is to make sure the destination location id starts changing as soon as possible, resulting in minimal number of steps to finish the setup for the warp.
- Profit.
Special Thanks
- primorial_soup and p4wn3r for the previous publications of the category and anyone else who contributed to those.
- CasualPokePlayer for the Blue no save corruption movie that inspired this one.
- Chamale for the help on TASing Pokemon Gen 1.
- YOU, for reading the submission text and watching the movie.
Last but not least, happy new year!
DrD2k9: This situation is similar to how
Gold/Silver vs
Crystal require different setups in order to perform similar glitches and to how
Red/Blue warrant separate publication vs
Yellow; because the setups required to achieve the "game end glitches" are unique as the fastest way to reach the game end in those releases. The Japanese versions of gen 1 Pokémon also have unique "game end glitch" setup (the Select Glitch, which iteslf has various uses) compared to the North American releases and thus warrant a separate publication to recognize the method reuqired for fastest non save corruption "game end glitch" use in Japanese releases.
This run does accomplish its goal faster than the most recently published
Green Version and would be acceptable from the standpoint of beating the 'current' publication. Unfortunately, within just a few days of this run being submitted (and before judgement could be rendered),
#9457: CasualPokePlayer's GB Pocket Monsters: Green Version "game end glitch" in 04:18.68 was also submitted. This latter run also uses the Select Glitch to accomplish a non save-corruption based "game end glitch run," and is a little faster. While the submitted branch labels are different (this submission being "warp glitch" and the other "game end glitch"), they are effectively the same goal (using a "game end glitch" to get to the Hall of Fame ASAP) and use the same base glitch (the Select Glitch) to accomplish this.
When we have two runs of the same game/goal on the workbench simultaneously and they are both acceptable compared to current publications of that game/goal, what actually happens to the two runs depends on the status of the prior run once then second run is submitted:
- When the earlier (assumed slower) submission has already been judged and accepted by the time the second (assumed faster) sumbission is submitted, then publication proceeds as normal for the slower run even though it is expected to be shortly obsoleted by the faster run.
- In other words, we're not going to retroactively recind an otherwise earned acceptance on a submission just because a new run happened to be submitted before the publication team was able to get the slower accepted run published.
- When the earlier (assumed slower) submission has not yet been judged/accepted by the time the second (assumed faster) sumbission is submitted, then the secondary faster submission is effectively proof to the judge(s) that the first submission is sub-optimal and thus not acceptable.
