Program Advance (PA)

The combo layer. Specific chips in specific letter sequences trigger unique fused super-attacks — and the recipes are largely undocumented in-game. The community recipe lists are part of the metagame.

How it works

Selecting 3 chips (or 5 in some recipes) at the chip-draw screen in a specific name + code pattern triggers a Program Advance instead of three individual attacks. The PA fires a single, much more powerful, often elaborate attack with bespoke animation.

Two main families:

Numerical PAs

Three chips of the same family in consecutive codes:

• CannonA + CannonB + CannonC → ZetaCannon (rapid burst of cannon shots)
• HiCannonA + HiCannonB + HiCannonC → ZetaHiCannon (stronger version)
• Sword + WideSword + LongSword*** of the same code → LifeSword (a giant sword that hits the entire enemy field in a + shape)

Same name, sequential codes → "stacking" PAs. Same family, mixed names with one shared code → "fusion" PAs.

Special PAs

Hidden recipes that don't follow the sequential-code pattern. Examples:

• GutsImpact + GutsShoot + GutsPunch (BN2) → super-Guts ultimate
• MasterStyle in BN5 — recipe-only ultimate combining specific signature chips
• BugCharge in later games

Special PAs are deliberately not telegraphed by the in-game UI. You discover them through:

• In-game NPC hints (occasionally)
• Magazine/strategy guide recipes (originally)
• Community wiki recipe lists (now)

What PAs do to deckbuilding

PAs are why folders look the way they do. A folder built around the LifeSword PA needs:

• 1× Sword of code A
• 1× WideSword of code A
• 1× LongSword of code A
• Plus 27 other chips that don't conflict with that code-A spine

So a "LifeSword folder" is built backwards from the recipe. The PA is the goal; the rest of the folder feeds the goal.

This converts deckbuilding into recipe-shaped optimization. You're not stacking damage cards in general — you're trying to make a specific 3-chip combo appear in the same opening Hand often enough that the PA is reliable.

What this teaches as a design pattern

PAs are the cleanest "hidden knowledge depth" pattern I've seen in any card-based game. The mechanic itself is in the manual; the list of recipes is genuinely emergent + community-curated.

This shape — the game ships a system, the community ships the canonical reference for it — is the same shape as PoE's community-driven loot filter culture. GGG ships the filter language; NeverSink ships the filter. Capcom shipped the PA system; community wikis ship the recipes.

The cost: new players don't discover most PAs by themselves. The benefit: experienced players have a constant learning surface — there's always one more recipe to find. Twenty years on, niche PA recipes are still being documented in retrospective guides.

BN5's one-PA-per-battle restriction

BN5 added a controversial limit: you can only fire one Program Advance per battle. This was meant to balance PAs that had become defining — by BN4, players who knew the recipes were one-shotting bosses with PA stacking. BN5's cap forced PA use into a "save it for the right moment" decision.

BN6 partially walked this back: multi-PA returned, but boss damage was rebalanced so PAs weren't an automatic win.

This is a textbook iterative-balance moment — see series evolution.

Patterns this exemplifies

• card-combo-recipes — multi-card combos that produce unique fused effects, distinct from sum-of-parts.
• hidden-knowledge-depth — recipes that aren't fully documented in-game; community knowledge is a meta-game layer.