Submission #6222: Wobmiar's GC Final Fantasy Crystal Chronicles "all dungeons" in 1:52:04.20

Final Fantasy Crystal Chronicles is an action role playing game developed by The Game Designers Studio and published by Square Enix in 2003 for the Nintendo GameCube. This was the first Final Fantasy game to include real-time fighting, which has since become a staple of Final Fantasy games. It was also the first Final Fantasy game to be published on a Nintendo console since Final Fantasy VI on the SNES.

You play as a caravanner, trying to beat dungeons to collect myrrh drops to be able to sustain your village's crystal energy, which protects the villagers from the mortal effects of the miasma. In this run, you go out as a Yuke, a race of mages, to go explore the whole world and beating every boss in the game to cleanse the world of all miasma corrupted monsters.

It's necessary to provide the game with a formatted memory card before being able to play the movie file.

This is my sixth TAS of this game, and fifth time on this category. I chose this category because with All Dungeons, we can see more of the game, the different dungeons and the bigger picture of the overworld. In any%, you do almost the same thing but you skip longer dungeons, which I don't skip here. Also, I do glitchless because the only glitches known in this game are multiplayer exclusive.

Final Fantasy is my favorite game of all time, even if it's mostly nostalgia. Here, you go on an epic quest to discover characters and cities, new stories unfold and there is sidequests offered to you along the way. Unfortunately, in this run, I skip most of it. In this game, the fastest way to beat it is to get to the last dungeon, wich is blocked by a miasma wall. The only way to pass it is to get the unknown element in the Lynari Desert, which is only accessable at year 5. You have to collect 3 drops of myrrh to complete a year, wich means I have to beat 12 dungeons before even being able to get to the last boss, 13 if we count reviving the flowers in Veo Lu Sluice to rise up the river to even get there. I do all 14 single dungeons in this game, and I also do a revisit of River Belle Path to get a necessary drop.

I chose to play the Yuke race, because he is not only the fastest caster, but his casts are even faster than autoattacks. Not only that, but spells are way more flexible and fusions can offer some great gameplay to beat boss easily. Also, with this race, I skip having to buy equipment, instead I focus on getting good artifacts to gain magic power. Since it's a TAS, I dont have to worry that much about my defense either since I can manipulate rngs all I want.

You choose, after each dungeon, an artifact that will give you permanent bonuses. In my case, I obviously choose the ones helpful for my magician. There is 2 ways to get artifacts: with points, or by dropping them inside the dungeon. The game works on a global timer, so each frame dungeons get a different seed wich affects bonus points conditions and ennemy drops (in the first room). The best way to manipulate what you want is to wait frames before entering a dungeon or a zone. It's also the same strategy I do after beating bosses to get the good set of artifacts I want. You also gain bonus points by killing ennemies and picking up things. It is important to know that my magical power only grow each multiple of 5. So most of the time I try to get to the higher one, with fruits and droppable artifacts. We start with 15, and we will be at the last boss with 75 (76 actually, but the 1 magic does not do any difference).

Money is also something important in this game, to pay the boat's fee. It costs me a total of 1200 gils, 2200 if I want to do a dialogue skip of 1 second. I got just over 1200 with all the money I dropped, and the items I sold to my merchant dad. I also take so detours here and there to change the element of my cristal, to be able to pass miasma streams.

This is an improvement of 2 minutes 8 seconds over the previous publication. First 3 years (9 dungeons) of the run are exactly the same imputs than the any% category.

New strategy on this first dungeon compared to the precedent run, since I decided to make some time sacrifice to gain more magic power, so I need 119 points. Taking 2 Blizzard Magicite and fusing them, plus the fire magicite for the Firaga spell, makes the boss faster than ever. (Kris (magic+3) is won.

The artifact Silver Bracer (+1) is taken from the chest, and I also pick the regular Fire Magicite from the hell plant. Some strategy occured to me to take time eating a carrot and eating a fruit to gain life and magic respectively to be able to beat the boss faster (and not needing to wait for Mog who's slowing at the end).

I drop a fire magicite, which is still the best element to deal damage to early bosses of the run. The Mage Masher (+1) is dropped from the troll, and, since we beat the boss faster than ever because of better optimizations and also better magic power. We have to note that in the japanese version, fruit effect lasts longer (and does not stop at a loading zone) so it's useful to beat the troll faster.

Here I want 110 points, wich is pretty easy since the boss have many minions. I also take advantage of that by picking up a fruit for later. I take the thunder magicite for damage and stun, and finish the level with the Wonder Wand Artifact (+3).

I take the bomb's key because it's faster than killing the worm. Also, as long as you have no fire magicite in you inventory, you can manipulate fire magicite to drop from the containers in the second part of the dungeon. Contrarily to the precedent TAS, I only take 2 magicites to make Fira since I did't take another command slot yet. Still, my magic is so high I beat the boss faster. I take the Dragon's Wisker artifact (magic+1).

I need 84 points at least to be able to manipulate the needed artifact, Red Slippers (magic+3). In the dungeon, I manipulate the switch to always have the Yuke tribe logo when needed, and since the fruit effet lasts, I don't have to take the bonus artifact in the chest to get to 25 magic. It decays at the 3/4 of the boss tho. Fire magicite is used.

In cycle 2, every ennemy is way more powerful than before. I need 119 points again to get the Mage's Staff (+5), and I take in my way 2 blizzard magicites and a fruit. Blizzara proves hitself very fast again to beat the Giant Crab.

I take the fast thunder magicite here, and go take the artifact Cat's Bell (magic+1) in the chest. I manipulate the bos to always do it's inhaling ability, until he's scripted to jump, and there I fuse Thundaga with Mog. A fruit is eaten for more damage.

We need to manipulate the content of the chest to get a blizzard magicite (1/4), then manipulate the position of the 2 golden lizards (1/12), and finally manipulate a mythril from the nearest mace lizardman (1/2). We don't need bonus points because we will get the Chocobo Pocket for 0 point, wich gives our third command slot. Mythril is our main source of money, giving 875 gils when sold to you father.

A blizzard magicite is manipulated on the first chest. I then proceed to kill 3 monsters, first a Coeurl to get a blizzard magicite to fusion them, and second the Ogre who drops the rtifact Noah's Lute (magic+5) at the same time as another Coeurl for a fruit.

This is a long and linear dungeon. We need to get fire and life magicite to get holy, the only thing making the boss vulnerable, and nothing else. The main goal here is to manipulate the monsters, particularly bombs and those annoying sludge monsters, to let me pass fast when they are blocking the route. We finish with the artifact Sage's Staff (magic+3).

I need all elements magicites in this dungeon to unlock doors, including life for holy. I kill the bat with simple attacks because it's faster, since he is flying. I make a really hard detour to take the Rune Bell artifact (magic+3). At the end I run blindly out of range of Mog, taking a carrot to survive. Against the Lich, we only need to destroy one switch to destroy her shield. Gravity destroys almost half her hp bar, so it's worth it.

I do all the complex steps of getting the unknown element as fast as possible, including skipping a menuing by making gravity with Mog. The artifact Book Of Light (magic+1) is taken from the chest, and we go at the boss with 3 blizzard magicites for a triple fusion. I kill every monsters in 1 cycle, and use normal combos to kill the stone scorpions. I can also hit the boss before it even starts with the help of Mog, with some precise timing.

Heal a flower, pick a key, run away. Repeat 3 times. I can get 2 fire magicites wich are super effective against the boss. The Blizzard Ring artifact (magic+1) gives you the equivalent of a free magicite of blizzard, wich is useful at the last dungeon.

I need a final amount of 75 magic power. So I drop +9, +9, +5 and +5, and I eat my final fruit (+2). I also get 2 other ice magicites, because the wide fusion is very important against the last boss (all that combined have a chance of 1/128 happening, hard rng manipulations were done) You have to wait for meteor parasite to get vulnerable, so I make some shenanigans in between cycles. Raem and Memiroa are manipulated to be as fast as possible.

When I started any%, I changed my early game to sacrifice some frames for better magic, and it turned out to be so effective I saved about 55 seconds on the first 3 years. Since those 3 years are exactly the same as all dungeons, I HAD to complete the run. I saved more than a minute for the last 2 years, so I'm pretty happy with that. It's also thanks for the japanese version wich makes the fruit effet last after loading zones.

ThunderAxe31: Judging.

ThunderAxe31: File replaced with a 138 blank frames trimmed version.

ThunderAxe31: Setting to delayed, in view of a potential improvement of the fastest-completion branch, which could indirectly affect the judgment for this submission.

ThunderAxe31: Judging resumed in view of a new fastest-completion submission.

ThunderAxe31: Despite the fact that we already have two "all dungeons" publications, I have come to the conclusion that movies made for this branch are not acceptable anymore, due of the recent fastest-completion submission.

Let's first summarize what happened before. When the currently published "all dungeons" movies (#5739, #5871) were submitted, there were no fastest-completion TASes to compare them with, and they were still faster than all the known RTA records at the time, including any%. Other than that, the goal itself of those movies were clear and definite, and the audience seemed to apprecciate their entertaning value; so I had to aknowledge that those movies were acceptable as a non-Vaultable branch, that is anything that can't be considered as fastest-completion (aka any%) nor full-completion (aka 100%).

The problem now is that a movie, in order to be accepted as a non-Vaultable branch, needs to feature differing contents compared to the fastest-completion and full-completion branches. As I already said above, at the time those "all dungeons" movies were submitted, there were no Vaultable branches to compare them with; however now we have a fastest-completion TAS submitted, for which I have to note how its contents are similar compared to "all dungeons" movies; more details about my comparisons can be found in this post. Also note that using a different character race didn't affect much the comparisons, despite the fact that I wanted to delay the judgment in light of such possibility.

For the reasons just explained, I'm rejecting this movie due of its goal choice.

Please note that this is not permanent change, as the situation may change again in case of future submissions introducing different movie contents. If, for example, a new strategy or a new glitch is discovered, whose utilization causes either of these branches to become enough different from each other, then the branches could be separated again, in light of the new contents introduced by such hypothetical submission. For the same reason, sometimes can happen the opposite: having different branches getting fused together due to a new submitted movie being too much similar to both of them, a situation commonly known as 'double obsoletion' (examples: #5291, #5833).