1
0
Fork 0
mirror of https://github.com/imjasonh/nescript synced 2026-07-08 00:45:38 +00:00

Add comprehensive documentation

docs/language-guide.md (895 lines):
  Complete reference for every language feature with code examples.
  Covers: program structure, types, expressions, statements, assets,
  mappers, error codes, and compiler flags.

docs/architecture.md (130 lines):
  Compiler pipeline overview, module descriptions, testing guide.

docs/nes-reference.md (190 lines):
  NES hardware quick reference: CPU, memory map, PPU, iNES format.

https://claude.ai/code/session_01W6eQFStA66EuMKHUFo2rx3
This commit is contained in:
Claude 2026-04-12 00:41:34 +00:00
parent b8c9e41276
commit d7092f703d
No known key found for this signature in database
3 changed files with 1215 additions and 0 deletions

130
docs/architecture.md Normal file
View file

@ -0,0 +1,130 @@
# Compiler Architecture
An overview of the NEScript compiler internals for contributors and maintainers.
---
## Pipeline
```
Source (.ne) --> Lexer --> Parser --> Analyzer --> IR Lowering --> Optimizer --> Codegen --> Assembler --> Linker --> ROM (.nes)
```
Each phase is a pure function (input to output) with no global state, making every stage independently testable.
```
.ne source ---------> Lexer -----> Parser -----> Analyzer ------>
(tokens) (AST) (annotated AST)
IR Lowering -----> Optimizer -----> Codegen -----> Linker -----> ROM
(IR) (optimized IR) (6502 insns) (.nes file)
```
### Phase Summary
| Phase | Input | Output | Responsibility |
|-----------------|----------------------|------------------------|------------------------------------------------------|
| **Lexer** | Source text | Token stream | Tokenization, number/string literal parsing |
| **Parser** | Token stream | AST | Syntax validation, tree construction |
| **Analyzer** | AST | Annotated AST | Type checking, scope resolution, call graph analysis |
| **IR Lowering** | Annotated AST | NEScript IR | Flatten expressions, expand u16 ops, desugar |
| **Optimizer** | IR | Optimized IR | Constant folding, dead code, ZP promotion, inlining |
| **Codegen** | Optimized IR | 6502 instruction list | Register allocation, instruction selection |
| **Assembler** | 6502 instructions | Byte sequences + fixups| Opcode encoding, address resolution |
| **Linker** | Bytes + assets | .nes file | Bank layout, vectors, iNES header |
---
## Modules
### `lexer/`
Tokenizes NEScript source text into a stream of typed tokens with source spans. Handles decimal, hex, and binary integer literals, string literals, all keywords, and operators.
### `parser/`
Recursive descent parser that converts the token stream into an AST. Defines all AST node types (expressions, statements, declarations) in `ast.rs`.
### `analyzer/`
Performs semantic analysis on the AST: type checking (`types.rs`), scope and symbol table management (`scope.rs`), and call graph construction with depth analysis (`call_graph.rs`). Detects recursion, type mismatches, undefined references, and call depth violations.
### `ir/`
Defines the intermediate representation and the lowering pass (`lowering.rs`) that translates the annotated AST into IR. Flattens nested expressions, expands 16-bit operations into 8-bit sequences, and resolves syntactic sugar.
### `optimizer/`
Runs optimization passes over the IR: constant folding (`const_fold.rs`), dead code elimination (`dead_code.rs`), zero-page promotion analysis (`zp_promote.rs`), and function inlining (`inliner.rs`).
### `codegen/`
Translates optimized IR into 6502 instructions. Includes register allocation for the A/X/Y registers (`regalloc.rs`) and instruction pattern matching for idiomatic 6502 code (`patterns.rs`).
### `asm/`
The built-in assembler. Encodes 6502 instructions (`encode.rs`) with all addressing modes (`addressing.rs`), using a complete opcode table (`opcodes.rs` -- 56 instructions across all modes).
### `linker/`
Assigns addresses to code and data segments, resolves fixups/relocations (`fixups.rs`), and handles bank allocation (`banks.rs`) for banked mappers.
### `rom/`
Builds the final iNES ROM file. Generates the 16-byte iNES header (`header.rs`) and places the NMI/RESET/IRQ vector table (`vectors.rs`).
### `runtime/`
Contains built-in runtime code that the compiler emits into every ROM: NES hardware initialization (`init.rs`), NMI handler generation (`nmi.rs`), controller read routines (`input.rs`), OAM DMA setup (`oam.rs`), PPU helper routines (`ppu.rs`), and software multiply/divide (`math.rs`).
### `assets/`
The asset pipeline. Converts PNG images to CHR tile data (`chr.rs`), generates nametables from full-screen images (`nametable.rs`), extracts and maps palettes (`palette.rs`), and handles audio import stubs (`audio.rs`).
### `debug/`
Debug instrumentation output. Generates source maps relating ROM addresses to source locations (`source_map.rs`), symbol tables compatible with Mesen (`symbols.rs`), and runtime check code for debug builds (`checks.rs`).
### `errors/`
Error reporting infrastructure. Defines the `Diagnostic` struct with error codes, severity levels, source spans, labels, help text, and notes (`diagnostic.rs`). Renders diagnostics with color and source context for terminal output (`render.rs`).
---
## Testing
### Test Organization
Tests are co-located with each module in `tests.rs` files:
```
src/lexer/tests.rs -- lexer unit tests
src/parser/tests.rs -- parser unit tests
src/analyzer/tests.rs -- semantic analysis tests
src/ir/tests.rs -- IR lowering tests
src/optimizer/tests.rs -- optimizer tests
src/codegen/tests.rs -- code generation tests
src/asm/tests.rs -- assembler tests
src/linker/tests.rs -- linker tests
src/rom/tests.rs -- ROM builder tests
src/assets/tests.rs -- asset pipeline tests
```
Integration tests live in the `tests/` directory:
```
tests/integration/ -- full pipeline tests with .ne source files
tests/error_tests/ -- tests that verify specific error codes
tests/asm_tests/ -- 6502 opcode and addressing mode tests
```
### Running Tests
```bash
# Run all tests
cargo test
# Run tests for a specific module
cargo test --lib lexer
cargo test --lib parser
cargo test --lib analyzer
# Run integration tests only
cargo test --test '*'
# Run a specific test by name
cargo test test_name
```
### Test Strategy
Each compiler phase is designed as a pure function, so unit tests provide isolated input and verify output without side effects. Integration tests compile complete `.ne` source files and verify the output ROM matches expected golden files in `tests/integration/expected/`.
Error tests in `tests/error_tests/` contain intentionally broken programs and verify that the correct error code is produced (e.g., `recursion.ne` should produce `E0402`).

895
docs/language-guide.md Normal file
View file

@ -0,0 +1,895 @@
# NEScript Language Guide
NEScript is a statically-typed, compiled language designed for NES game development. It compiles directly to 6502 machine code packaged as iNES-format ROMs -- no external assembler or tooling required.
This guide covers every language feature with practical examples.
---
## Program Structure
Every NEScript program consists of a game declaration, top-level definitions, and a start declaration.
```
game "My Game" {
mapper: NROM
mirroring: vertical
}
const SPEED: u8 = 2
var score: u8 = 0
fun helper() -> u8 {
return 42
}
state Title {
on frame {
draw Logo at: (100, 100)
if button.start {
transition Playing
}
}
}
state Playing {
on enter {
score = 0
}
on frame {
// game logic here
}
}
start Title
```
### Game Declaration
The `game` block is required and must appear first. It names the game and sets hardware configuration.
```
game "Coin Cavern" {
mapper: NROM
mirroring: vertical
}
```
Available properties:
| Property | Values | Default |
|--------------|----------------------------------|--------------|
| `mapper` | `NROM`, `MMC1`, `UxROM`, `MMC3` | required |
| `mirroring` | `horizontal`, `vertical` | `horizontal` |
### Start Declaration
Exactly one `start` declaration must exist. It names the initial state entered on power-on.
```
start Title
```
---
## Types
NEScript has four primitive types and fixed-size arrays.
### Primitive Types
| Type | Size | Range | Description |
|--------|---------|-----------------|------------------------------------|
| `u8` | 1 byte | 0 to 255 | Unsigned 8-bit integer |
| `i8` | 1 byte | -128 to 127 | Signed 8-bit integer |
| `u16` | 2 bytes | 0 to 65535 | Unsigned 16-bit integer |
| `bool` | 1 byte | `true` / `false`| Boolean |
### Arrays
Arrays are fixed-size, homogeneous, and zero-indexed. The size must be a compile-time constant. Maximum 256 elements.
```
var enemies: u8[8]
const TABLE: u8[4] = [10, 20, 30, 40]
```
### Type Casting
NEScript has no implicit coercion. All conversions use `as`:
```
var a: u8 = 200
var b: u16 = a as u16 // zero-extend: 200
var c: i8 = a as i8 // reinterpret bits
var d: u8 = b as u8 // truncate to low byte
```
---
## Variables
### Variable Declarations
Variables are declared with `var` and must have an explicit type:
```
var x: u8 // uninitialized (zeroed on state entry)
var y: u8 = 100 // initialized
var pos: u16 = 0x0400 // 16-bit value
var alive: bool = true
var scores: u8[4] = [0, 0, 0, 0]
```
### Constants
Constants are evaluated at compile time and stored in ROM:
```
const MAX_ENEMIES: u8 = 5
const SPEED: u8 = 3
const SIN_TABLE: u8[8] = [0, 49, 90, 117, 127, 117, 90, 49]
```
### Memory Placement Hints
The NES has 256 bytes of zero-page RAM with faster access. You can hint where variables should be placed:
```
fast var px: u8 // prefer zero-page (faster instructions)
slow var high_score: u16 // prefer upper RAM (saves zero-page space)
var normal: u8 // compiler decides automatically
```
If zero-page is exhausted and `fast` variables cannot be placed, the compiler emits error `E0301`.
### Scope
| Scope | Declared In | Lifetime |
|----------|----------------|---------------------------------------------|
| Global | Top level | Entire program, permanent RAM allocation |
| State | `state` block | Active while state is active; RAM reusable |
| Function | `fun` block | Duration of function call |
| Block | `if`/`while` | Enclosing block, shares parent allocation |
---
## Functions
### Declaration
Functions use `fun`, with optional parameters and return type:
```
fun add(a: u8, b: u8) -> u8 {
return a + b
}
fun reset_score() {
score = 0
}
```
### Inline Functions
The `inline` keyword hints the compiler to inline the function at call sites:
```
inline fun clamp(val: u8, max: u8) -> u8 {
if val > max {
return max
}
return val
}
```
`inline` is a hint -- the compiler may decline for large functions.
### Calling Functions
```
var result: u8 = add(10, 20)
reset_score()
```
### Restrictions
- **No recursion.** Both direct and indirect recursion are compile errors (`E0402`).
- **Call depth limit.** The default maximum call depth is 8. Exceeding it produces error `E0401`.
---
## States
States are the top-level organizational unit. Exactly one state is active at any time.
### State Declaration
```
state Playing {
var timer: u8 = 0 // state-local variable
on enter {
// runs once when entering this state
timer = 60
}
on exit {
// runs once when leaving this state
}
on frame {
// runs every frame (60 Hz) while this state is active
timer -= 1
draw Player at: (player_x, player_y)
}
}
```
`on frame` is syntactic sugar for a loop with an implicit `wait_frame()` at the end. A state can have any combination of `on enter`, `on exit`, and `on frame`.
### State Transitions
```
transition GameOver
```
Transitions are immediate. The current state's `on exit` runs, then the target state's `on enter` runs. The remainder of the current frame handler does not execute.
---
## Expressions
### Literals
```
42 // decimal integer
0xFF // hexadecimal
0b10110001 // binary
1_000 // underscores allowed for readability (if supported)
true // boolean
false // boolean
[1, 2, 3] // array literal
```
All integer literals must fit in `u16` (0-65535). The compiler narrows to the required type at usage.
### Arithmetic Operators
| Operator | Description | Example |
|----------|----------------|--------------|
| `+` | Addition | `a + b` |
| `-` | Subtraction | `a - b` |
| `*` | Multiplication | `a * b` |
| `/` | Division | `a / b` |
| `%` | Modulo | `a % b` |
`*`, `/`, and `%` are available but expensive on the 6502 (software routines). The compiler optimizes power-of-two operations to shifts and warns on non-power-of-two multiply/divide.
### Bitwise Operators
| Operator | Description | Example |
|----------|----------------|--------------|
| `&` | Bitwise AND | `a & 0x0F` |
| `\|` | Bitwise OR | `a \| 0x80` |
| `^` | Bitwise XOR | `a ^ mask` |
| `~` | Bitwise NOT | `~a` |
| `<<` | Shift left | `a << 2` |
| `>>` | Shift right | `a >> 1` |
### Comparison Operators
| Operator | Description | Example |
|----------|-------------------|--------------|
| `==` | Equal | `a == 0` |
| `!=` | Not equal | `a != b` |
| `<` | Less than | `a < 10` |
| `>` | Greater than | `a > max` |
| `<=` | Less or equal | `a <= 255` |
| `>=` | Greater or equal | `a >= min` |
### Logical Operators
NEScript uses keyword-based logical operators:
```
if alive and (health > 0) {
// ...
}
if not paused or force_update {
// ...
}
```
| Operator | Description |
|----------|---------------|
| `and` | Logical AND |
| `or` | Logical OR |
| `not` | Logical NOT |
### Operator Precedence
From highest to lowest:
| Level | Operators | Associativity |
|-------|------------------------------------|---------------|
| 1 | `()` grouping | -- |
| 2 | `-` (unary), `~`, `not` | right |
| 3 | `*`, `/`, `%` | left |
| 4 | `+`, `-` | left |
| 5 | `<<`, `>>` | left |
| 6 | `&` | left |
| 7 | `^` | left |
| 8 | `\|` | left |
| 9 | `==`, `!=`, `<`, `>`, `<=`, `>=` | left |
| 10 | `and` | left |
| 11 | `or` | left |
### Button Reads
Read controller input as boolean expressions:
```
if button.right {
player_x += SPEED
}
if button.a {
jump()
}
```
Available buttons: `up`, `down`, `left`, `right`, `a`, `b`, `start`, `select`.
For two-player games, prefix with the player:
```
if p1.button.a { /* player 1 */ }
if p2.button.right { /* player 2 */ }
```
Without a prefix, `button` refers to player 1.
### Function Calls in Expressions
```
var clamped: u8 = clamp_x(player_x + SPEED)
```
### Array Indexing
```
var val: u8 = table[i]
table[i] = 0
```
### Type Casting
```
var wide: u16 = narrow as u16
```
---
## Statements
### Assignment
```
x = 10
x += 5
x -= 1
x &= 0x0F
x |= 0x80
x ^= mask
```
All assignment operators:
| Operator | Description |
|----------|---------------------|
| `=` | Assign |
| `+=` | Add and assign |
| `-=` | Subtract and assign |
| `&=` | AND and assign |
| `\|=` | OR and assign |
| `^=` | XOR and assign |
Array element assignment:
```
enemies[i] = 0
scores[player] += 10
```
### If / Else If / Else
Braces are always required. No ternary operator.
```
if health == 0 {
transition GameOver
} else if health < 3 {
flash_warning()
} else {
// normal gameplay
}
```
### While Loop
```
var i: u8 = 0
while i < 10 {
enemies[i] = 0
i += 1
}
```
### Loop (Infinite)
```
loop {
wait_frame()
if button.start {
break
}
}
```
The compiler warns if a `loop` contains neither `break`, `wait_frame`, nor `transition`.
### Break and Continue
```
var i: u8 = 0
while i < 20 {
i += 1
if enemies[i] == 0 {
continue // skip inactive enemies
}
if i > 10 {
break // stop processing
}
update_enemy(i)
}
```
### Return
```
fun abs_diff(a: u8, b: u8) -> u8 {
if a > b {
return a - b
}
return b - a
}
```
Functions without a return type use `return` with no value (or simply reach the end of the function body).
### Draw
Render a sprite to the screen:
```
draw Player at: (player_x, player_y)
draw Coin at: (COIN_X, COIN_Y) frame: anim_frame
```
The `draw` statement writes to the OAM shadow buffer. The NES supports up to 64 sprites per frame.
Syntax: `draw SpriteName at: (x_expr, y_expr) [frame: expr]`
### Transition
Switch to another state immediately:
```
transition GameOver
```
The current state's `on exit` runs, then the target state's `on enter` runs.
### Wait Frame
Yield execution until the next vertical blank (NMI). Synchronizes to the 60 Hz display refresh.
```
wait_frame()
```
This triggers OAM DMA transfer and PPU updates before yielding. Inside `on frame`, a `wait_frame()` is implicit at the end of each frame.
### Scroll
Set the PPU scroll position:
```
scroll(scroll_x, scroll_y)
```
### Load Background
Load a background nametable:
```
load_background TitleBG
```
### Set Palette
Apply a palette:
```
set_palette GamePalette
```
### Function Calls as Statements
```
reset_score()
update_physics(player_x, player_y)
```
---
## Assets
### Sprite Declarations
```
sprite Player {
chr: @chr("assets/player.png")
}
sprite Coin {
chr: @binary("assets/coin.bin")
}
```
### Palette Declarations
Define color palettes using NES PPU color indices (`$00`-`$3F`):
```
palette GamePalette {
colors: [0x0F, 0x00, 0x10, 0x30,
0x0F, 0x07, 0x17, 0x27,
0x0F, 0x09, 0x19, 0x29,
0x0F, 0x01, 0x11, 0x21]
}
```
Each group of 4 values is a sub-palette. The first color is typically `0x0F` (black) as the shared background color.
### Background Declarations
```
background TitleBG {
chr: @chr("assets/title_screen.png")
}
```
### Asset Sources
Three ways to provide asset data:
| Source | Description |
|----------------------------|---------------------------------------|
| `@chr("file.png")` | Convert PNG to CHR tile data |
| `@binary("file.bin")` | Include raw binary data verbatim |
| Inline `[0x00, 0x7E, ...]`| Hex byte array directly in source |
---
## Mappers
The mapper determines cartridge hardware and available ROM size.
| Mapper | PRG ROM | CHR ROM | Features |
|---------|---------------|----------------|----------------------------------|
| `NROM` | 16 or 32 KB | 8 KB | No banking, simplest |
| `MMC1` | Up to 256 KB | Up to 128 KB | Switchable banks |
| `UxROM` | Up to 256 KB | 8 KB CHR RAM | PRG banking only |
| `MMC3` | Up to 512 KB | Up to 256 KB | Scanline counter, banking |
### Bank Declarations
For mappers with bank switching:
```
bank MainCode {
// Always-resident code (NMI handler, core engine)
}
bank Level1 {
state Level1 { ... }
background Level1BG { ... }
}
```
Banks can hold `prg` (code/data) or `chr` (graphics) content. Transitions between states in different banks automatically emit bank-switch and trampoline code.
---
## Comments
```
// Line comment -- extends to end of line
/* Block comment
spans multiple lines */
```
---
## Includes
Split your game across multiple files:
```
include "physics.ne"
include "enemies.ne"
```
Includes are resolved relative to the including file. Circular includes are a compile error. Duplicate includes are skipped automatically.
---
## Debug Mode
Compile with `--debug` to enable runtime instrumentation. All debug features are stripped completely in release builds (zero bytes, zero cycles).
### Debug Logging
```
debug.log("Player position: ", px, ", ", py)
```
### Debug Assertions
```
debug.assert(lives > 0, "Lives should never be negative")
```
### Runtime Checks (Debug Only)
In debug mode, the compiler inserts:
- Array bounds checking on indexed access
- Arithmetic overflow warnings
- Stack depth monitoring at function entry
- Frame overrun detection (warns if frame handler exceeds vblank period)
---
## Inline Assembly
For performance-critical code, drop to 6502 assembly:
### Bound Assembly
```
fun fast_shift(input: u8) -> u8 {
asm {
lda {input}
asl a
asl a
sta {return}
}
}
```
`{variable_name}` resolves to the variable's memory address. `{return}` is the return value location.
### Raw Assembly
```
raw asm {
.org $C000
nop
rti
}
```
Raw blocks bypass all compiler management. Use with extreme caution.
---
## Error Codes
### Lexer Errors (E01xx)
| Code | Description |
|--------|----------------------------|
| E0101 | Unterminated string literal |
| E0102 | Invalid character |
| E0103 | Number literal overflow |
### Type Errors (E02xx)
| Code | Description |
|--------|----------------------------|
| E0201 | Type mismatch |
| E0202 | Invalid cast |
| E0203 | Invalid operation for type |
### Memory Errors (E03xx)
| Code | Description |
|--------|----------------------------|
| E0301 | Zero-page overflow |
### Control Flow Errors (E04xx)
| Code | Description |
|--------|----------------------------|
| E0401 | Call depth exceeded |
| E0402 | Recursion detected |
| E0403 | Unreachable state |
| E0404 | Transition to undefined state |
### Declaration Errors (E05xx)
| Code | Description |
|--------|----------------------------|
| E0501 | Duplicate declaration |
| E0502 | Undefined variable |
| E0503 | Undefined function |
| E0504 | Missing start declaration |
| E0505 | Multiple start declarations|
### Warnings (W01xx)
| Code | Description |
|--------|------------------------------------------|
| W0101 | Expensive multiply/divide operation |
| W0102 | Loop without break or wait_frame |
| W0103 | Unused variable |
| W0104 | Unreachable code |
### Example Error Output
```
error[E0201]: type mismatch
--> game.ne:42:15
|
42 | var x: u8 = -5
| ^^ expected u8, found negative integer
|
= help: use i8 if you need negative values: var x: i8 = -5
```
```
error[E0402]: recursion is not allowed
--> game.ne:55:5
|
55 | flood_fill(x + 1, y)
| ^^^^^^^^^^^^^^^^^^^^
|
= note: flood_fill calls itself (directly recursive)
= help: the NES has only 256 bytes of stack; use an iterative algorithm instead
```
---
## Compiler Commands
### Build
Compile a `.ne` source file into a `.nes` ROM:
```
nescript build game.ne
nescript build game.ne --output my_game.nes
nescript build game.ne --debug
nescript build game.ne --asm-dump
```
| Flag | Description |
|---------------|------------------------------------------------|
| `--output` | Set output ROM file path (default: input.nes) |
| `--debug` | Enable debug mode with runtime checks |
| `--asm-dump` | Dump generated 6502 assembly to stdout |
### Check
Type-check a source file without producing a ROM:
```
nescript check game.ne
```
---
## Complete Example
A full game demonstrating states, input, functions, constants, and transitions:
```
game "Coin Cavern" {
mapper: NROM
}
const SPEED: u8 = 2
const SCREEN_RIGHT: u8 = 240
const COIN_X: u8 = 180
const COIN_Y: u8 = 100
var player_x: u8 = 40
var player_y: u8 = 200
var score: u8 = 0
var coins_left: u8 = 3
fun clamp_x(val: u8) -> u8 {
if val > SCREEN_RIGHT {
return 0
}
return val
}
state Title {
on frame {
draw Logo at: (100, 100)
if button.start {
transition Playing
}
}
}
state Playing {
on enter {
player_x = 40
player_y = 200
score = 0
coins_left = 3
}
on frame {
if button.right {
player_x += SPEED
if player_x > SCREEN_RIGHT {
player_x = SCREEN_RIGHT
}
}
if button.left {
if player_x >= SPEED {
player_x -= SPEED
} else {
player_x = 0
}
}
if player_x >= COIN_X {
if player_y >= COIN_Y {
score += 1
coins_left -= 1
if coins_left == 0 {
transition GameOver
}
}
}
draw Player at: (player_x, player_y)
draw Coin at: (COIN_X, COIN_Y)
}
}
state GameOver {
on frame {
draw Trophy at: (120, 100)
if button.start {
transition Title
}
}
}
start Title
```
Build and run:
```
nescript build coin_cavern.ne
# produces coin_cavern.nes -- open in any NES emulator
```

190
docs/nes-reference.md Normal file
View file

@ -0,0 +1,190 @@
# NES Hardware Quick Reference
A concise reference to the NES hardware for NEScript contributors. Understanding these constraints explains many of the compiler's design decisions.
---
## CPU: Ricoh 2A03 (MOS 6502 variant)
- **Clock speed:** 1.79 MHz (NTSC), 1.66 MHz (PAL)
- **Architecture:** 8-bit data bus, 16-bit address bus
- **Registers:**
- `A` (Accumulator) -- 8-bit, used for arithmetic and logic
- `X` (Index X) -- 8-bit, used for indexing and counting
- `Y` (Index Y) -- 8-bit, used for indexing and counting
- `SP` (Stack Pointer) -- 8-bit, points into the $0100-$01FF range
- `P` (Status) -- 8-bit flags: N(egative), V(overflow), B(reak), D(ecimal), I(nterrupt), Z(ero), C(arry)
- `PC` (Program Counter) -- 16-bit
- **No multiply, divide, or floating point instructions**
- **Instruction set:** 56 official opcodes with multiple addressing modes (implied, immediate, zero page, zero page X/Y, absolute, absolute X/Y, indirect, indexed indirect, indirect indexed)
### Why This Matters for NEScript
The 6502's 3-register architecture drives the compiler's register allocator. The lack of multiply/divide means the compiler emits software routines for `*`, `/`, and `%`. Zero-page addressing is 1 byte shorter and 1 cycle faster than absolute addressing, which is why `fast` variable placement matters.
---
## Memory Map
| Address Range | Size | Description |
|-------------------|------------|-------------------------------------|
| `$0000`-`$00FF` | 256 bytes | Zero page -- fast access |
| `$0100`-`$01FF` | 256 bytes | Stack (grows downward) |
| `$0200`-`$02FF` | 256 bytes | OAM shadow buffer (convention) |
| `$0300`-`$07FF` | 1280 bytes | General purpose RAM |
| `$0800`-`$1FFF` | -- | Mirrors of $0000-$07FF |
| `$2000`-`$2007` | 8 bytes | PPU registers |
| `$2008`-`$3FFF` | -- | Mirrors of PPU registers |
| `$4000`-`$4017` | 24 bytes | APU and I/O registers |
| `$4018`-`$401F` | 8 bytes | CPU test mode (normally disabled) |
| `$4020`-`$5FFF` | -- | Expansion ROM (mapper-dependent) |
| `$6000`-`$7FFF` | 8 KB | SRAM / PRG RAM (if present) |
| `$8000`-`$BFFF` | 16 KB | PRG ROM lower bank |
| `$C000`-`$FFFF` | 16 KB | PRG ROM upper bank (or fixed bank) |
### Interrupt Vectors
| Address | Vector | Description |
|----------|---------|------------------------------------------|
| `$FFFA` | NMI | Non-Maskable Interrupt (vertical blank) |
| `$FFFC` | RESET | Power-on / reset entry point |
| `$FFFE` | IRQ | Interrupt Request (mapper-dependent) |
### NEScript Memory Usage
The compiler reserves the following:
- `$00`-`$0F`: System use (frame counter, temp registers, NMI flags)
- `$10`-`$FF`: Available for `fast` variables and compiler-promoted variables
- `$0200`-`$02FF`: OAM shadow buffer (DMA'd to PPU each frame)
- `$0300`-`$07FF`: General variables, state-local storage
---
## PPU (Picture Processing Unit)
- **Resolution:** 256 x 240 pixels
- **Refresh rate:** 60 Hz (NTSC), 50 Hz (PAL)
- **VRAM:** 2 KB internal (2 nametables), expandable by mapper
- **Pattern tables:** 2 tables of 256 tiles each (one for backgrounds, one for sprites), stored in CHR ROM/RAM
- **Tile size:** 8x8 pixels (or 8x16 for tall sprites)
- **Color depth:** 2 bits per pixel (4 colors per tile, selected from a sub-palette)
### Nametables
- 4 logical nametables, each 960 bytes of tile indices + 64 bytes of attribute data
- With 2 KB VRAM, 2 physical nametables exist; the other 2 are mirrors
- Mirroring arrangement (horizontal or vertical) is set by the cartridge/mapper
### Sprites (OAM)
- **64 sprites** total, each defined by 4 bytes:
- Byte 0: Y position
- Byte 1: Tile index
- Byte 2: Attributes (palette, priority, flip H/V)
- Byte 3: X position
- **8 sprites per scanline** maximum (excess sprites are dropped)
- OAM is 256 bytes, typically shadow-buffered at CPU `$0200` and DMA'd via `$4014`
### Palettes
- **Background:** 4 sub-palettes, each with 4 colors (first color shared across all)
- **Sprite:** 4 sub-palettes, each with 4 colors (first color is transparent)
- Colors are indices into the NES master palette (64 entries, `$00`-`$3F`)
- Palette RAM is 32 bytes total at PPU `$3F00`-`$3F1F`
### PPU Registers
| Address | Name | Description |
|---------|----------|--------------------------------------|
| `$2000` | PPUCTRL | NMI enable, sprite size, pattern table select, nametable select |
| `$2001` | PPUMASK | Color emphasis, sprite/background enable, clipping |
| `$2002` | PPUSTATUS| Vblank flag, sprite 0 hit, sprite overflow |
| `$2003` | OAMADDR | OAM address for writes |
| `$2004` | OAMDATA | OAM data read/write |
| `$2005` | PPUSCROLL| Scroll position (write twice: X, Y) |
| `$2006` | PPUADDR | VRAM address (write twice: high, low)|
| `$2007` | PPUDATA | VRAM data read/write |
### Rendering Timing
- **Vblank** starts at scanline 241 and lasts ~20 scanlines (~2,273 CPU cycles)
- PPU updates (palette, nametable, scroll) must happen during vblank
- `on frame` code runs during the visible frame; the implicit `wait_frame()` yields until the next vblank
- Approximately **29,780 CPU cycles per frame** (NTSC)
---
## APU (Audio Processing Unit)
| Channel | Type | Description |
|-----------|------------|------------------------------------|
| Pulse 1 | Square wave| Variable duty cycle (12.5/25/50/75%) |
| Pulse 2 | Square wave| Same as Pulse 1 |
| Triangle | Triangle | Fixed volume, good for bass |
| Noise | Noise | Pseudo-random, for percussion |
| DMC | Sample | 1-bit delta-modulated samples |
APU registers span `$4000`-`$4017`. The audio driver (included automatically when `sfx` or `music` declarations exist) runs during NMI.
---
## iNES ROM Format
The standard ROM file format for NES emulators:
```
Offset Size Description
------ ------- -----------
0 4 bytes Magic number: "NES" followed by $1A
4 1 byte PRG ROM size in 16 KB units
5 1 byte CHR ROM size in 8 KB units
6 1 byte Flags 6: mapper (low nibble), mirroring, battery, trainer
7 1 byte Flags 7: mapper (high nibble), VS/Playchoice, NES 2.0
8-15 8 bytes Padding (zeros)
16+ varies PRG ROM data (N x 16384 bytes)
after varies CHR ROM data (N x 8192 bytes)
```
### Common Mapper Numbers
| Number | Name | PRG ROM | CHR | Notes |
|--------|--------|-------------|-------------|----------------------------|
| 0 | NROM | 16/32 KB | 8 KB ROM | No bank switching |
| 1 | MMC1 | Up to 256 KB| Up to 128 KB| Switchable 16 KB PRG banks |
| 2 | UxROM | Up to 256 KB| 8 KB RAM | Switchable 16 KB PRG banks |
| 4 | MMC3 | Up to 512 KB| Up to 256 KB| Scanline counter, 8 KB banks|
### Mirroring
- **Horizontal:** nametables A-A-B-B (vertical scrolling games)
- **Vertical:** nametables A-B-A-B (horizontal scrolling games)
- Set in byte 6, bit 0 of the iNES header (0 = horizontal, 1 = vertical)
---
## Controller
Each controller has 8 buttons read as a serial shift register via `$4016` (port 1) and `$4017` (port 2).
Read sequence:
1. Write `$01` then `$00` to `$4016` to strobe (latch button states)
2. Read `$4016` (or `$4017`) 8 times; bit 0 of each read is one button
Button order: A, B, Select, Start, Up, Down, Left, Right.
The NEScript runtime handles this automatically. The programmer reads button state via `button.a`, `button.up`, etc.
---
## Key Constraints for NEScript
| Constraint | Limit | NEScript Response |
|-------------------------|----------------|------------------------------------------|
| Total RAM | 2 KB | Static allocation, no heap |
| Zero page | 256 bytes | `fast`/`slow` hints, compiler promotion |
| Stack | 256 bytes | Call depth limit, no recursion |
| Sprites per frame | 64 | Compiler manages OAM buffer |
| Sprites per scanline | 8 | Hardware limit, no workaround |
| Vblank time | ~2,273 cycles | PPU updates must be fast |
| Frame budget | ~29,780 cycles | Frame overrun detection in debug mode |
| No multiply/divide HW | -- | Software routines, power-of-2 optimization|