The Conde Petie Mountain Path

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.

A climb up through tree roots between Conde Petie and the coast. It is short and there is exactly one thing to keep track of.

The four coloured stones

There are four statues on the way up, each holding a stone.

• Blue — right-hand fork of the first screen, near the Remedy
• Red — second statue, up the vine stairs, near the Tent
• Yellow — only reachable after you kill Hilgigars
• Green — the lower fork near the Ether

That last one is the one people miss. Two guides go out of their way to warn about it.

Once you have all four, put them into the statue near the moogles and you get a Moonstone. If you are chasing the achievement for collecting every Moonstone in the game, this is one of them.

The chests

Remedy, Tent and Ether, one on each of three screens. The Tent here is a chest and is not the same Tent Stiltzkin sells you — several guides list both and it is easy to tick off the wrong one.

Stiltzkin, at 666 gil

Magic Tag, Tent, Ether. He is near Suzuna.

This is his fourth stop. The payoff for buying from him is cumulative across all eight, so if you are going for it, do not skip this one because money is tight.

Suzuna wants that letter

If you took Mogmatt’s letter from the Conde Petie Item Shop, Suzuna is who it is for. She gives you a Kupo Nut.

Catch the Oglop

There is a bug on the path. Run at it and hammer the confirm button.

Take it back to Bryan Rootrunner in Conde Petie and he gives you the Oglop Card. It needs a walk back, which is why most people never do it.

Hilgigars

8,106 HP. It has no elemental weakness, so do not go hunting for one — Ramuh is not the answer here despite what one guide says.

Slow and Silence both land, and that is the actual strategy. Steal a Phoenix Down and a Mythril Fork while you are at it.

Game Notes

This is a fetch quest with a very short leash, and it is a good one.

Four stones, four statues, all on the road you are already walking, and the only one that takes a decision is the Green — down a fork you have no reason to take after the boss is already dead and the path onward is obvious.

That is the whole design. The reward is real, the collection is invisible unless you are looking, and nothing about it is ever explained to you. You walk past a statue, notice it has a stone in it, and either start paying attention or don’t.

FFIX does this a lot and rarely better than here.

Completionist mode. Full spoilers, in play order, with every permanent loss called out before you reach it. 3 entries below are marked where our sources disagree with each other — we have not picked a winner for you.

Ids 300–303 are the four Coloured Stones

We closed ids 304–316 last area as the thirteen Stellazzio, in zodiac order. Directly beneath that block our data showed four consecutive unnamed key items — 300, 301, 302, 303 — all awarded on this path and nowhere else.

Nine publishers say there are exactly four coloured stones here. Blue, Red, Yellow, Green, traded at a statue for a Moonstone. Nothing else in the area is a key item, and nothing is left over on either side.

So the key-item table now reads, continuously:

Two adjacent blocks, seventeen ids, fully named. That is the first stretch of that table we can read straight through.

It also came cheap, and it is worth being clear about why: we did not decode anything here. We noticed four consecutive unknowns in one area and asked whether the published record contained a set of four collectables in that area. It did, unanimously. The decode was done by the guides; our data only supplied the question.

Hilgigars does not carry the +10,000 offset

8,106 HP at level 28, confirmed by six independent publishers — Caves of Narshe, RPGClassics, Game8, Gamer Guides, GameFAQs and ffexodus. Our stored figure is 8,106. Nobody prints 18,106.

This matters because Hilgigars is unambiguously a boss: a scripted story fight, zero experience, its own battle scene. And in the same batch as this, the Scarlet Hair record at Madain Sari splits the corpus exactly on a leading digit.

So the offset is not a property of boss records. It is a property of some records, and Hilgigars is a clean control sitting right next to a dirty one.

We spent part of this batch trying to find what separates them — a flag, a level band, a scene property — and found a candidate that looked perfect across ten hand-picked records and died on the full table within a minute. That is written up in our notes rather than here, because a hypothesis that survived four minutes is not a finding. What matters for a reader is the conclusion: we still cannot tell from the data alone which records are affected, so we still publish the figures the guides publish.

The weakness that is not there

Six publishers say Hilgigars has no elemental weakness. So does our field data — the weakness bitmask reads zero, which is rare and which we would normally treat with suspicion rather than confidence.

ffexodus says it is weak to thunder and recommends bringing Ramuh.

One against six, with our own data on the majority’s side, and the dissent carries a strategy recommendation attached — which is the version of being wrong that costs a reader something. Somebody following that advice spends MP on a summon for no effect and concludes the fight is harder than it is.

Game8 adds, alone, that Hilgigars is immune to holy and shadow and absorbs healing magic. Nobody confirms or denies it. Our bitmasks read zero for immunity, which is consistent with “no elemental immunity” and says nothing about healing, since absorb-heal is a separate property we do not extract. Carried as a single-source detail rather than contradicted.

The real answer to the fight is Slow and Silence, which Gamer Guides gives and which is a status question, not an elemental one.

The stone exchange, and whether it repeats

One publisher states flatly that the exchange can only be done once. Another’s wording implies it repeats as long as stones are available. No third source addresses it.

We are reporting the disagreement rather than picking, because the practical advice is identical either way: collect all four, trade them once, take the Moonstone. Nobody claims a second Moonstone from this statue, so the only thing the dispute could cost you is a hypothetical you have no reason to attempt.

Three publishers tie the Moonstone to the collect-them-all achievement in the re-releases, which is the reason to bother at all.

Two Tents, and the Stiltzkin signature

Six publishers list a Tent in a chest and a Tent in Stiltzkin’s bundle as separate things, and they are right — our field data shows both.

That is worth a note about our own table. This batch we confirmed that all eight of Stiltzkin’s bundles appear in our award data as ordinary item awards, three at a time, at an exact 250-byte spacing, with literal counts. They look exactly like three chests in a row.

Our earlier rule for spotting a purchase — a blank count, meaning the price was a variable — does not catch these, because Stiltzkin’s prices are fixed numbers written into the script. A fixed-price purchase is indistinguishable from a chest by that test. The 250-byte triple is the signature that does catch them.

We had already got Burmecia, Cleyra and Fossil Roo right, because publishers told us Stiltzkin was there. Had we been relying on the blank-count rule, we would have shipped twenty-four phantom chests across eight areas.

Everything else, and one lone claim

Ochu 3,568 HP weak to fire, Gnoll 1,375 dropping a Sapphire, Troll 1,469 — all three match our data exactly against two independent publishers each. The Gnoll’s card really is Mimic, which looked like a reader error on our side and is not, for the third time in two areas.

Suzuna is here, she is who Mogmatt’s letter is for, and she pays a Kupo Nut — three publishers name the reward. The Oglop is on the path and the card trade is back in Conde Petie; five publishers document both halves and none of them routes you back for it in the walkthrough text, which is why almost nobody does it.

One publisher lists a loose Phoenix Pinion “in a corner by stairs”. Ten do not, our field data does not carry it, and a Phoenix Pinion is an Ochu steal on this path — which is the obvious way a steal becomes a pickup in someone’s notes. Filed as a single-source claim we do not believe rather than dropped silently, because a reader who has seen that guide deserves to know we looked.

Game Notes

The coloured stones landed in our laps and the reason is worth naming.

We have spent this project treating the published corpus as something to check our data against, and most of the interesting results have come from our data seeing something the guides missed — a refill nobody wrote down, an area with two halves, a boss record ten thousand off.

This one went the other way round. Our data could see four consecutive unnamed key items on one path and could not tell you a single thing about what they were. Nine guides could, instantly and unanimously, because a person walked up the path and looked at the statues.

Neither source is the authority. The data knows the shape of the table; the guides know what is in the world. The decode needed both, and it took about ten minutes once the question was asked the right way round.