Mythril Mine and Fort Condor

Story mode. Written for a first playthrough and kept free of spoilers — it will not tell you what happens here, only what the game is asking of you.

Two things here, and they could not be less alike: a cave you walk through in fifteen minutes, and a strategy game bolted onto the side of a mountain.

Climb the vines

There is a Materia up there called Long Range, and it is one of the best things you will find in this half of the game.

It does one thing: a character carrying it hits for full damage from the back row. Which means a sword user can stand at the back, take half damage, and give up nothing — the same deal Barret has had for free since the first reactor.

Take it, put it on whoever is most fragile, and forget about it. You will notice the difference for the next twenty hours.

It is easy to miss because it is up a climbable wall rather than on the path, and because nothing in the cave suggests you should be looking up.

Everything else in the mine is optional

There are chests and a couple of floor pickups. The mine stays open all game, so none of it needs doing now.

There is a scene partway through with some people in dark suits. It is not a fight. Let it happen.

Fort Condor

The mountain with the reactor on top, just past the mine. Climb it and an old man asks for help defending it.

Here is the honest version: this is a whole separate strategy game, it runs for most of FF7, and you can ignore essentially all of it. You buy units with your own gil and place them to stop soldiers climbing a hill. If you like that sort of thing, it is genuinely good and it pays out consistently.

If you do not, skip it. There is exactly one battle here that matters and it happens much later, and it does not require you to have played any of the others.

What you should do on this visit is talk to the old man and say yes — the conversation is worth points on the hidden counter regardless of whether you play — and then decide for yourself.

Game Notes

Long Range is a quiet masterpiece of an item and almost nobody talks about it.

FF7’s row system is one of its better ideas and one of its least explained. The back row halves what you take and halves what you deal, except for weapons the game has decided are long-ranged, which lose nothing. It never tells you which weapons those are. It never tells you that a Materia exists to change the answer. It puts that Materia up a wall in a cave, on the way to a boss that hovers out of melee reach.

That is the game’s whole teaching method in one sequence: here is a system, here is the tool that breaks it open, here is a fight that punishes you for not having found either. No tutorial, no prompt, no arrow. Just a wall you can climb if you think to look.

Fort Condor, meanwhile, is FF7 at its most 1997 — an entire second genre stapled to the side of a mountain because somebody on the team wanted to make it, running for thirty hours, almost entirely optional, and with one hard consequence hidden right at the end of it. Nobody would ship that now. The game is better for it.

Completionist mode. Full spoilers, in play order, with every permanent loss called out before you reach it. 4 entries below are marked where our sources disagree with each other — we have not picked a winner for you.

Spoilers from here.

The mine, quickly

Long Range Materia, up the vines. Full damage from the back row for whoever holds it, which converts a melee character into someone who can sit at the back taking half damage for nothing. The next boss hovers out of melee reach, so the game introduces the solution and the problem within about forty minutes of each other.

Then an Ether, a Tent, a Mind Source, a Hi-Potion and an Elixir scattered across chests and floors. Two guides list the same six pickups but disagree about which screens the Elixir and Hi-Potion sit on. The mine never closes, so it does not matter much.

Grand Glove is a steal from Madouge at 12.7%. Only one guide documents it, and the game’s own tables confirm it — Madouge’s ordinary drop is a Hi-Potion at the same rate, and it morphs into one too, so the Glove is the only reason to care about the encounter. It is 31 attack at 110 accuracy with four slots, against the 24 of the Mythril Claw Tifa is probably still wearing.

Flamethrower comes off the Ark Dragons here, and it is worth stopping for. It is the strongest thing an Enemy Skill Materia holds for a while, and the boss two areas from now has no elemental weakness for anything else to exploit. The dragons absorb Fire, nullify Earth and take double from Wind, so do not fight fire with fire.

Keep Steal equipped while you farm the skill. Ark Dragon carries an Ether at 50.8% — a coin flip, the best steal rate in the mine, off an enemy that lives here permanently. Its drops are a Phoenix Down at 12.7% and a second Ether at 3.2%, so robbing it is roughly four times better than killing it.

The Turks turn up in the western section. It is not a fight. Elena’s debut, and she immediately gives away where Sephiroth is going, which is the scene’s only purpose. One guide says you get a dialogue choice here; no source documents any consequence to it.

Fort Condor

Here is the part where the record falls apart.

Fort Condor is a real-time tower defence game attached to the side of a mountain. You buy units with your own gil, place them above a line, and stop Shinra soldiers climbing to the reactor. It recurs at intervals for most of FF7, gets harder as you win, and ends in a single mandatory battle on Disc 2 with a permanent consequence.

What is settled

• You need 4,000 gil on you to start a battle at all.
• Winning on the field returns 200 gil per surviving unit.
• If soldiers reach the top, you drop into a normal RPG fight against CMD. Grand Horn.
• Battle Level rises every three wins, and each level has a one-time prize — Magic Comb, then Peace Ring, then Megalixir, then Superball — plus a repeatable consumable for every win after that. What the prizes actually are: the Magic Comb is 37 attack on double growth with three unlinked slots; the Peace Ring blocks Sadness, Fury, Confusion and Berserk, four statuses off one accessory; and the Superball is 68 attack with no growth and no Materia slots at all, which is the one nobody warns you about — it earns no AP and holds no magic, so the raw number is the whole of it.
• You can decline and let the fort fight alone for 3,000 gil of its own money. It scores as a win and advances your level, but pays no prize.
• Everything you donate is handed straight back to you immediately before the final battle. Donating is a loan, not a cost.

What is not settled

How many battles exist. One source says seventeen and numbers them differently from everyone else. Three say twenty scheduled plus a mandatory final. The reward-to-battle mapping is offset by one or two battles between the two fullest schedules. Neither has been reconciled by anyone.

Whether the skip system exists at all. Three sources document the 3,000-gil delegation in detail. A fourth documents the minigame thoroughly and does not mention it.

How the battles are scheduled. One source explains it through an internal plot-progression counter — a battle unlocks when the counter rises by six and is marked missed when it rises by seventy past that — and gives a full table of values. Nobody else describes the mechanism at all. It is the only account that explains why several battles are structurally impossible, and it is single-source.

The Grand Horn rule, which everyone states loosely

Beating CMD. Grand Horn after losing the field counts as a win in the first battle only. At every other level it counts as neither — no prize, no gil, no penalty. You keep his drop and his gil and nothing else moves.

Except at the very end, where it counts as a win again. Which produces the single most useful thing on this page.

The final battle

Mandatory, Disc 2, cannot be delegated. Win and you get the Phoenix Materia — which spawns on the mountaintop and has to be walked up to and collected, it is not handed over — plus the Huge Materia.

Lose it and the fort closes permanently. Phoenix is gone. There is no second attempt. That is the only hard permanent loss in thirty hours of this sidequest, and it sits at the very end of it.

No prior wins are required. You can ignore Fort Condor completely for the entire game and still take the Phoenix, provided you win this one fight.

Deliberately losing is better than winning

Settled from the game’s own tables. CMD. Grand Horn exists in three versions: 2,000 HP dropping a Vagyrisk Claw, 4,000 HP dropping nothing, and 8,000 HP dropping the Imperial Guard every single time. The gil tracks the HP — 2,400, 4,800 and 9,600 — so the version everyone is trying to avoid is also the one that pays four times the others.

The 8,000 HP version is the one you meet in the final battle. Beating him counts as winning it. So letting the tower-defence phase fail and killing him yourself gets you the Phoenix, the Huge Materia, a guaranteed Imperial Guard and 9,600 gil — and you spend nothing on units getting there.

The armour is not a consolation prize. 82 defence, 74 magic defence, and six slots in three linked pairs, against 46, 28 and four on the Gold Armlet. Win the field cleanly and you never see it at all.

What is actually missable here

• Phoenix Materia and Huge Materia — lost forever if you lose the final battle.
• The one-time level prizes (Magic Comb, Peace Ring, Megalixir, Superball) — each is awarded once, at a different Battle Level, and you only reach higher levels by winning. Run out of playable battles before reaching Level 4 and the Superball is gone.
• Imperial Guard — only by losing the final field phase and winning the boss fight.
• Everything in the mine — not missable. It stays open all game.

Game Notes

Fort Condor is the clearest case in FF7 of a system nobody fully documented, and the reason is structural rather than lazy: it is scheduled off an invisible counter that advances with story progress, so no two players see the same battles at the same times, and a guide author writing from one playthrough is writing from a sample of one.

That is why the schedules disagree by one or two battles, why the counts differ by three, and why one of the four sources does not know the skip system exists. It is not that somebody was careless. It is that the thing is genuinely hard to observe from the inside, and nobody went and read it out of the game until much later.

The practical upshot is small and slightly absurd. Thirty hours of tower defence, one real consequence, and the optimal play at the moment of consequence is to lose on purpose.