Two sprite boxes can overlap while their visible shapes still miss each other.

Use it in a project
P0–P7 and Q0–Q7 are eight MSB-left mask rows. Place Q SHIFT_X pixels right and SHIFT_Y pixels down relative to P. Return HIT=1 and the first overlapping P row when any opaque bits coincide; otherwise HIT=0 and ROW=255. A shift of eight or more on either axis is a valid empty overlap. Inputs are preserved.
Include component.inc once (MOS callers include the shared 6502 implementation through it). Reserve the public fields in this exact order: HIT, ROW, SHIFT_X, SHIFT_Y, P0, P1, P2, P3, P4, P5, P6, P7, Q0, Q1, Q2, Q3, Q4, Q5, Q6, Q7. They occupy 20 contiguous bytes. Also reserve the private workspace: T0, T1, T2, T3, T4, T5 (6 bytes). Array names in the explanation refer to the numbered consecutive bindings, such as C0, C1 and so on.
Call pixel_mask_collision_init once at the intended reset boundary. It clears inputs as well as outputs and workspace. Set your starting state and configuration afterwards, then call pixel_mask_collision_step whenever the described event occurs. The helper reads no device, waits for no frame and installs no interrupt handler.
| Ownership | Fields |
|---|---|
| Retain between calls | none |
| Supply before service | SHIFT_X, SHIFT_Y and both mask arrays |
| Read after service | HIT and ROW |
Shift each Q row into P’s horizontal alignment, then AND corresponding rows. Any surviving bit is a shared opaque pixel. Transparent corners of overlapping boxes no longer count as contact.
Reusable initialisation and service
; SPDX-License-Identifier: MIT
; Test Two Small Opaque Pixel Masks. See README.md for the byte interface and ownership.
; init clears the public fields and private workspace; set inputs afterwards.
pixel_mask_collision_init:
ld a,0
ld (HIT),a
ld (ROW),a
ld (SHIFT_X),a
ld (SHIFT_Y),a
ld (P0),a
ld (P1),a
ld (P2),a
ld (P3),a
ld (P4),a
ld (P5),a
ld (P6),a
ld (P7),a
ld (Q0),a
ld (Q1),a
ld (Q2),a
ld (Q3),a
ld (Q4),a
ld (Q5),a
ld (Q6),a
ld (Q7),a
ld (T0),a
ld (T1),a
ld (T2),a
ld (T3),a
ld (T4),a
ld (T5),a
ret
pixel_mask_collision_step:
; HIT = 0
ld a,0
ld (HIT),a
; ROW = 255
ld a,255
ld (ROW),a
; if SHIFT_X >= 8:
ld a,(SHIFT_X)
cp 8
jp nc,pixel_mask_collision_L1
jp pixel_mask_collision_L2
pixel_mask_collision_L1:
; return
ret
jp pixel_mask_collision_L3
pixel_mask_collision_L2:
pixel_mask_collision_L3:
; if SHIFT_Y >= 8:
ld a,(SHIFT_Y)
cp 8
jp nc,pixel_mask_collision_L4
jp pixel_mask_collision_L5
pixel_mask_collision_L4:
; return
ret
jp pixel_mask_collision_L6
pixel_mask_collision_L5:
pixel_mask_collision_L6:
; T0 = 0
ld a,0
ld (T0),a
; T5 = 8 - SHIFT_Y
ld a,8
ld b,a
ld a,(SHIFT_Y)
ld c,a
ld a,b
sub c
ld (T5),a
; while T0 < T5:
pixel_mask_collision_L7:
ld a,(T0)
ld b,a
ld a,(T5)
ld c,a
ld a,b
cp c
jp c,pixel_mask_collision_L8
jp pixel_mask_collision_L9
pixel_mask_collision_L8:
; T1 = Q[T0]
ld a,(T0)
ld e,a
ld d,0
ld hl,Q0
add hl,de
ld a,(hl)
ld (T1),a
; T2 = SHIFT_X
ld a,(SHIFT_X)
ld (T2),a
; while T2 != 0:
pixel_mask_collision_L10:
ld a,(T2)
cp 0
jp nz,pixel_mask_collision_L11
jp pixel_mask_collision_L12
pixel_mask_collision_L11:
; T1 = T1 >> 1
ld a,(T1)
srl a
ld (T1),a
; T2 = T2 - 1
ld a,(T2)
sub 1
ld (T2),a
jp pixel_mask_collision_L10
pixel_mask_collision_L12:
; T3 = T0 + SHIFT_Y
ld a,(T0)
ld b,a
ld a,(SHIFT_Y)
ld c,a
ld a,b
add a,c
ld (T3),a
; T4 = P[T3]
ld a,(T3)
ld e,a
ld d,0
ld hl,P0
add hl,de
ld a,(hl)
ld (T4),a
; T4 = T4 & T1
ld a,(T4)
ld b,a
ld a,(T1)
ld c,a
ld a,b
and c
ld (T4),a
; if T4 != 0:
ld a,(T4)
cp 0
jp nz,pixel_mask_collision_L13
jp pixel_mask_collision_L14
pixel_mask_collision_L13:
; HIT = 1
ld a,1
ld (HIT),a
; ROW = T3
ld a,(T3)
ld (ROW),a
; return
ret
jp pixel_mask_collision_L15
pixel_mask_collision_L14:
pixel_mask_collision_L15:
; T0 = T0 + 1
ld a,(T0)
add a,1
ld (T0),a
jp pixel_mask_collision_L7
pixel_mask_collision_L9:
ret
component_end:Machine contract
AF/BC/DE/HL may change; IX/IY are preserved. A direct CALL needs two return-address stack bytes. The helper makes no nested calls and adds no other stack use. Its named scratch is private workspace, not a pointer, a ROM variable or shared zero-page space that the game may borrow concurrently. The samples use fixed symbol bindings. Scope or rename every label and binding, including the component_end measurement marker, and allocate a separate block for each independent instance. Different helpers can also use common field names, so combining them needs separate namespaces or explicitly renamed bindings. Give each block one mutation owner.
The 48K diagnostic boots through ROM tape loading and USR, disables interrupts and owns the screen. The helper performs no port or screen writes.
Adapt the caller
demo.asm supplies the bindings and include order. exercise.inc first runs table cases, then a separate client retaining state between calls. A game can call the public step entry directly and remove the diagnostic packet and display. Do not copy the diagnostic’s hardware takeover into an existing game loop.
Build with make ASM198X=/path/to/asm198x. The Makefile produces normal, custom, deliberately wrong and repaired programs. Tools must already be installed; the sample downloads nothing.
Bindings and include order
org $8000
CUSTOM equ 0
BROKEN equ 0
COYOTE_STEPS equ 4
BUFFER_STEPS equ 4
TEXT_CAP equ 8
PROPERTY_COUNT equ 16
STREAM_BYTES equ 13
CRC_POLY equ 7
include "caller.inc"
include "exercise.inc"
component_start:
include "component.inc"
include "data.inc"
include "cases.inc"
include "../../physics/rectangle-overlap/display.inc"
code_end:
STATE:
HIT: defb 0
ROW: defb 0
SHIFT_X: defb 0
SHIFT_Y: defb 0
P0: defb 0
P1: defb 0
P2: defb 0
P3: defb 0
P4: defb 0
P5: defb 0
P6: defb 0
P7: defb 0
Q0: defb 0
Q1: defb 0
Q2: defb 0
Q3: defb 0
Q4: defb 0
Q5: defb 0
Q6: defb 0
Q7: defb 0
T0: defb 0
T1: defb 0
T2: defb 0
T3: defb 0
T4: defb 0
T5: defb 0
input_ptr: defw 0
output_ptr: defw 0
cases_left: defb 0
ready: defb 0
preview: defs 8,0
OUTPUT: defs 1300,0
end startA client retaining state between calls
preview_rules:
if CUSTOM
; SHIFT_X = 0
ld a,0
ld (SHIFT_X),a
; P0 = 128
ld a,128
ld (P0),a
; Q0 = 128
ld a,128
ld (Q0),a
if BROKEN
; SHIFT_X = 0
ld a,0
ld (SHIFT_X),a
endif
call checked_step
call emit_state
ret
else
; SHIFT_X = 1
ld a,1
ld (SHIFT_X),a
; P0 = 128
ld a,128
ld (P0),a
; Q0 = 128
ld a,128
ld (Q0),a
if BROKEN
; SHIFT_X = 0
ld a,0
ld (SHIFT_X),a
endif
call checked_step
call emit_state
endif
retCustomise it
Align the two opaque pixels to produce an actual hit. The maintained custom caller changes its inputs or client sequence rather than hiding a different implementation behind the same label. Compare every state of the two clients, not only their final display.
Custom client selection
org $8000
CUSTOM equ 1
BROKEN equ 0
COYOTE_STEPS equ 2
BUFFER_STEPS equ 2
TEXT_CAP equ 4
PROPERTY_COUNT equ 16
STREAM_BYTES equ 13
CRC_POLY equ 49
include "caller.inc"
include "exercise.inc"
component_start:
include "component.inc"
include "data.inc"
include "cases.inc"
include "../../physics/rectangle-overlap/display.inc"
code_end:
STATE:
HIT: defb 0
ROW: defb 0
SHIFT_X: defb 0
SHIFT_Y: defb 0
P0: defb 0
P1: defb 0
P2: defb 0
P3: defb 0
P4: defb 0
P5: defb 0
P6: defb 0
P7: defb 0
Q0: defb 0
Q1: defb 0
Q2: defb 0
Q3: defb 0
Q4: defb 0
Q5: defb 0
Q6: defb 0
Q7: defb 0
T0: defb 0
T1: defb 0
T2: defb 0
T3: defb 0
T4: defb 0
T5: defb 0
input_ptr: defw 0
output_ptr: defw 0
cases_left: defb 0
ready: defb 0
preview: defs 8,0
OUTPUT: defs 1300,0
end startThe custom diagnostic row is 01 00 00 00 80 00 00 00. The full client trace in the source explains how it gets there.
Mistake and repair
The wrong caller ignores horizontal displacement and aligns the masks. It reports contact between opaque bits that are actually apart. Convert relative coordinates before the narrow-phase mask query.
The BROKEN branch lives in the sequential client, not the reusable helper. wrong.asm selects it; repaired.asm restores the normal call sequence. The repaired executable matches the normal program (for Spectrum, compare loaded CODE rather than the tape filename).
Select the deliberately wrong client
org $8000
CUSTOM equ 0
BROKEN equ 1
COYOTE_STEPS equ 4
BUFFER_STEPS equ 4
TEXT_CAP equ 8
PROPERTY_COUNT equ 16
STREAM_BYTES equ 13
CRC_POLY equ 7
include "caller.inc"
include "exercise.inc"
component_start:
include "component.inc"
include "data.inc"
include "cases.inc"
include "../../physics/rectangle-overlap/display.inc"
code_end:
STATE:
HIT: defb 0
ROW: defb 0
SHIFT_X: defb 0
SHIFT_Y: defb 0
P0: defb 0
P1: defb 0
P2: defb 0
P3: defb 0
P4: defb 0
P5: defb 0
P6: defb 0
P7: defb 0
Q0: defb 0
Q1: defb 0
Q2: defb 0
Q3: defb 0
Q4: defb 0
Q5: defb 0
Q6: defb 0
Q7: defb 0
T0: defb 0
T1: defb 0
T2: defb 0
T3: defb 0
T4: defb 0
T5: defb 0
input_ptr: defw 0
output_ptr: defw 0
cases_left: defb 0
ready: defb 0
preview: defs 8,0
OUTPUT: defs 1300,0
end start
Cost and alternatives
The measured helper extent is 282 code bytes; public state is 20 bytes and private scratch is 6 bytes. There are no immutable helper tables. Caller data, the test cases, return-address stack, alignment and display buffers cost additional space. These are assembled sizes, not cycle measurements.
At most eight row visits and seven shifts per row. Run an AABB broad phase first. Pre-shifted masks use more memory to reduce shifts; a box or circle may already provide the intended game feel.
This bounded helper covers Q right/down of P. Negative relative placement needs a deliberate swapped/normalised query, including conversion of the returned row to the desired space. It tests collision masks, not display colours or hardware collision latches.
Execution evidence
Four variants run on 48K PAL ROM tape load and USR, using Asm198x 0.0.72, Emu198x 0.44.2. Each executes 50 boundary/varied/error table cases, dirty initialisation and the separate retained-state client. An independent mathematical/container model checks every public byte after every call. This is bounded example coverage, not exhaustive correctness for every possible state.
The records retain loaded immutable code, completion, register sentinels, full packets, source/tool/firmware hashes and captures. The public gate validates these retained observations and rejects altered evidence; it does not run restricted firmware. No firmware is distributed. The eight displayed bytes are the first eight public fields, zero-padded when shorter; the packet includes every field.
Where it fits
Rectangle overlap: Use the cheap box test to avoid mask work for separated objects, then pay for precision only where it changes the game.