Gofus Retro

See what an item is worth broken, next to what it costs to craft

For items with characteristics, the price graph now draws a third line: what the runes from breaking one unit are worth, day by day, using your own saved rune prices. It is each rune's average yield multiplied by that rune's price, summed across the whole table, so it moves as the rune market moves. The legend now carries two margins side by side — Sell, the market value minus the craft cost, and Break, the rune value minus the craft cost — so you can tell at a glance whether a recipe is better sold or broken, and hovering the chart shows both for that day. As with the craft line, days without a recorded rune price reuse the last one you entered, and the line only appears once every rune in the item's table has a known price, so a missing rune price can never make breaking look less profitable than it really is. Items without characteristics do not break into runes and are unchanged.

The price graph now opens on items you have never priced

Until now the graph showed an empty state whenever you had no saved price for the item itself, even when everything needed to chart it was already there. For an item nobody sells and everybody breaks, the market value was never the interesting number anyway — the rune yield against the craft cost is. Those two lines now draw on their own. With no market line to lead, the rune yield takes its place as the main curve, the craft cost stays underneath it as the dashed line, and the legend shows the Break margin on its own, since there is no sale to compare against. If neither the recipe nor the runes can be priced from what you have saved, the graph still says so. Items you have priced are unchanged.

Rune breaking rates were too optimistic — now corrected

It started with your reports. Several of you told me the rune chances Gofus showed when breaking an item didn't match what you were actually getting. You were right — and confirming it took 640,940 broken items. Over the past weeks I broke 640,940 items on the beta server, every one at maximum stats, across 24 different item-and-rune combinations from level 40 up to level 195. Comparing what Gofus predicted against what really dropped showed the site was too optimistic — but only inside a specific band of item levels. Outside that band, the old numbers were already exact. Where the numbers moved, for a PA rune: · Below level 67 — nothing changed. Not by a single point. · Between level 67 and level 138 — the site was too generous, and the gap widens the higher you go. The peak sits right in the middle, around level 119, where 66.66% becomes 52.4%. That is the worst case anywhere in the game: 14 points off, about a fifth of the number displayed. Nothing was ever off by more than that. · Above level 138 — nothing changed either. The rate is capped at 66.66%, and always was. The band slides with the rune's weight: for a PM rune it runs from level 63 to 129 (worst around 111), for a PO rune from level 44 to 91 (worst around 78). Same shape, same ceiling on the error. The reassuring part is how narrow the damage is. This really only concerns the PA, PM and PO runes. Those come from a single point of stat, so the item's level is the only thing setting their rate — and the level part is precisely what was wrong. The everyday stat runes are untouched: Strength, Chance, Agility, Intelligence, Vitality, Wisdom and Prospecting were already exact on any real item, at any level. A few others — Critical, Summon, Damage — shift a little, but only on items below roughly level 65. In practice: level 100 boots with 1 PM go from 54% to 44%, a level 111 amulet with 1 PA from 58% to 47%, a level 124 amulet from 67% to 56%. The worst run I measured was a pair of level 110 boots — the site claimed 65% for a PM rune, and 15,000 breaks returned 51%. Not everything went down. The same pass fixed a separate bug that was hiding Ra runes: items whose stat rolls high enough to produce one were showing no chance of it at all. Ares and Epis de Shika both reach the Ra Cha threshold on a good Chance roll, and both displayed a flat zero. They now show a real chance of one, and every stat that reaches a rune threshold is fixed the same way. Your breaking estimates will look worse than they did yesterday, but they now match what actually comes out of the machine — which is the number you want before committing 10,000 items to it. And all of this happened because a few people took the time to say "this number looks wrong". If a rate on Gofus doesn't match what you see in game, please tell me: the item, the rune, and a rough count is plenty to get started. It is the only way these get found.