diff --git a/c64/day01.asm b/c64/day01.asm
index d5c7b46..2717b5e 100644
--- a/c64/day01.asm
+++ b/c64/day01.asm
@@ -1,122 +1,160 @@
 // vim: filetype=kickass
 
-*=* "Day 01"
+* = * "Day 01"
+
+
+//
+// day01
+//
+// Destroys:
+// a, x, y, $02..$04, $10..$1c, $20..$25, $fd..$fe, everything you've ever loved
+//
+.const input_pointer  = $03
+.const buffer_pointer = $20
+.const atoi_result    = $22
+
+.const outer_pointer  = $20
+.const inner_pointer  = $22
+.const remainder      = $24
+
+// Need to be $03 and $fd due to multiply_16bit_unsigned
+.const outer_value    = $03
+.const inner_value    = $fd
+
 day01:
-    move_16_imm($03, $04, str_title)
+    move_16_imm($03, str_title)
     jsr write_string
 
-    move_16_imm($03, $04, day01_input)
-    move_16_imm($25, $26, buffer)
+    //
+    // Parse the input at day01_input into an array of integers.
+    //
+    move_16_imm(input_pointer, day01_input)
+    move_16_imm(buffer_pointer, buffer)
     ldy #0
 
 !line:
-    lda #0
-    sta $22
-    sta $23
+    // Create a temporary 16-bit integer to store the result in
+    // in between digits.
+    move_16_imm(atoi_result, 0)
 
 !digit:
-    lda ($03), y
+    // Check for the newline character at the end of the line.
+    lda (input_pointer), y
     cmp #'\n'
-    beq !break+
+    beq !newline+
 
-    i16_mul10($22, $23)
+    // Multiply the stored number by 10.
+    i16_mul10(atoi_result)
 
-    lda ($03), y
+    // Subtract '0' from the current digit to convert it to "binary".
+    lda (input_pointer), y
     sec
     sbc #'0'
 
-    i16_i8_add_a($22, $23)
+    // Add the binary digit to the stored number
+    i16_i8_add_a(atoi_result)
 
+    // Increment y.
     iny
     bne !digit-
 
-    inc $04
+    // If y overflows, increment the MSB of the input pointer as well,
+    // for a 16-bit increment.
+    inc input_pointer + 1
     jmp !digit-
 
-!break:
-    sty $24
+!newline:
+    sty zp_temp
 
+    // Append the result of the conversion to the array.
     ldy #0
-    lda $22
-    sta ($25), y
+    lda atoi_result + 0
+    sta (buffer_pointer), y
+
     iny
-    lda $23
-    sta ($25), y
+    lda atoi_result + 1
+    sta (buffer_pointer), y
 
+    // Increment the buffer pointer by 2.
     lda #2
-    i16_i8_add_a($25, $26)
+    i16_i8_add_a(buffer_pointer)
 
-    ldy $24
+    ldy zp_temp
     iny
     bne !+
-    inc $04
+    inc input_pointer + 1
 !:
-    lda ($03), y
+
+    // Check for the null terminator.
+    lda (input_pointer), y
     bne !line-
 
-    move_16_imm($25, $26, buffer)
-
-    move_16_imm($03, $04, str_parse_done)
+    move_16_imm($03, str_parse_done)
     jsr write_string
 
+    move_16_imm(outer_pointer, buffer)
+
     // Add every number to every other number
 !outer:
-    move_16_imm($27, $28, buffer)
+    move_16_imm(inner_pointer, buffer)
 
     // $03..$04 = buffer[i]
     ldy #0
-    lda ($25), y
-    sta $03
+    lda (outer_pointer), y
+    sta outer_value
     iny
-    lda ($25), y
-    sta $04
+    lda (outer_pointer), y
+    sta outer_value + 1
 
     dey
-    lda ($25), y
+    lda (outer_pointer), y
     bne !not_zero+
     iny
-    lda ($25), y
+    lda (outer_pointer), y
     bne !not_zero+
     .break
     jmp !error+
 !not_zero:
 
     // Subtract from 2020 to see what number we need
-    i16_imm_i16_sub($fb, $fc, 2020, $03, $04)
+    i16_imm_i16_sub(remainder, 2020, outer_value)
 
 !inner:
     // $fd..$fe = buffer[j]
     ldy #0
-    lda ($27), y
-    sta $fd
+    lda (inner_pointer), y
+    sta inner_value
     iny
-    lda ($27), y
-    sta $fe
+    lda (inner_pointer), y
+    sta inner_value + 1
 
     dey
-    lda ($27), y
+    lda (inner_pointer), y
     bne !not_zero+
     iny
-    lda ($27), y
+    lda (inner_pointer), y
     bne !not_zero+
 
-    i16_inc($25, $26)
+    i16_inc(outer_pointer)
     jmp !outer-
-!not_zero:
 
+!not_zero:
     // j++
     lda #2
-    i16_i8_add_a($27, $28)
-    i16_i16_cmp_bne($fb, $fc, $fd, $fe, !inner-)
+    i16_i8_add_a(inner_pointer)
+
+    // Continue the loop if the number didn't match the needed number.
+    i16_i16_cmp_bne(remainder, inner_value, !inner-)
 
 !done:
     // Multiply the results together
     sec
     jsr multiply_16bit_unsigned
 
-    move_16_imm($03, $04, str_calc_done)
+    move_16_imm($03, str_calc_done)
     jsr write_string
 
+    // Print the 32-bit result as decimal.
     lda $25
     sta udivmod32_dividend + 3
     lda $24
@@ -130,7 +168,7 @@ day01:
     rts
 
 !error:
-    move_16_imm($03, $04, str_no_result)
+    move_16_imm($03, str_no_result)
     jsr write_string
     rts
 
diff --git a/c64/main.asm b/c64/main.asm
index aca4b8c..cccb712 100644
--- a/c64/main.asm
+++ b/c64/main.asm
@@ -38,11 +38,11 @@ day01_input:
 .const cursor_pointer_lo = $05
 .const cursor_pointer_hi = $06
 
-.macro move_16_imm(dst_lo, dst_hi, src) {
-    lda #<src
-    sta dst_lo
-    lda #>src
-    sta dst_hi
+.macro move_16_imm(dst, imm) {
+    lda #<imm
+    sta dst
+    lda #>imm
+    sta dst + 1
 }
 
 //
diff --git a/c64/math.inc b/c64/math.inc
index 7f61353..6a2b017 100644
--- a/c64/math.inc
+++ b/c64/math.inc
@@ -86,17 +86,17 @@
     i16_mul2(lo, hi)
 }
 
-// Destroys: a, lo, hi, $fd, $fe
-.macro i16_mul10(lo, hi) {
-    i16_mul2(lo, hi)
+// Destroys: a, $fd, $fe
+.macro i16_mul10(dst) {
+    i16_mul2(dst, dst + 1)
 
-    lda lo
+    lda dst
     sta $fd
-    lda hi
+    lda dst + 1
     sta $fe
 
-    i16_mul4(lo, hi)
-    i16_i16_add(lo, hi, $fd, $fe)
+    i16_mul4(dst, dst + 1)
+    i16_i16_add(dst, dst + 1, $fd, $fe)
 }
 
 .macro u32_mul2(lsb) {
@@ -140,13 +140,13 @@
     sta dst + 3
 }
 
-// Destroys: a, dst_lo, dst_hi
-.macro i16_i8_add_a(dst_lo, dst_hi) {
+// Destroys: a
+.macro i16_i8_add_a(dst) {
     clc
-    adc dst_lo
-    sta dst_lo
+    adc dst
+    sta dst
     bcc !cc+
-    inc dst_hi
+    inc dst + 1
 !cc:
 }
 
@@ -162,9 +162,9 @@
 }
 
 // Destroys: a, dst_lo, dst_hi
-.macro i16_inc(lo, hi) {
+.macro i16_inc(dst) {
     lda #1
-    i16_i8_add_a(lo, hi)
+    i16_i8_add_a(dst)
 }
 
 .macro i16_i16_sub_imm(dst_lo, dst_hi, src_lo, src_hi, imm) {
@@ -191,24 +191,24 @@
     sta dst_hi
 }
 
-.macro i16_imm_i16_sub(dst_lo, dst_hi, imm, src_lo, src_hi) {
-    lda src_lo
+.macro i16_imm_i16_sub(dst, imm, src) {
+    lda src
     eor #$ff
     sec
     adc #<imm
-    sta dst_lo
+    sta dst
 
-    lda src_hi
+    lda src + 1
     eor #$ff
     adc #>imm
-    sta dst_hi
+    sta dst + 1
 }
 
-.macro i16_i16_cmp_bne(a_lo, a_hi, b_lo, b_hi, target) {
-    lda a_lo
-    cmp b_lo
+.macro i16_i16_cmp_bne(a, b, target) {
+    lda a
+    cmp b
     bne target
-    lda a_hi
-    cmp b_hi
+    lda a + 1
+    cmp b + 1
     bne target
 }
diff --git a/c64/screen.asm b/c64/screen.asm
index eea294e..2370ddd 100644
--- a/c64/screen.asm
+++ b/c64/screen.asm
@@ -101,7 +101,7 @@ print_carriage_return:
 //  - $03-$04 - Pointer to null-terminated string.
 //  - $05-$06 - Cursor pointer.
 //
-// Destroys: a, x, y
+// Destroys: a, x, y, $02
 //
 write_string:
     ldx #0
@@ -118,7 +118,7 @@ write_string:
 
     sta (cursor_pointer_lo, x)
 
-    i16_inc(cursor_pointer_lo, cursor_pointer_hi)
+    i16_inc(cursor_pointer_lo)
 
     iny
     jmp !loop-
@@ -151,11 +151,11 @@ print_hex_0x:
 
     lda #'0'
     sta (cursor_pointer_lo, x)
-    i16_inc(cursor_pointer_lo, cursor_pointer_hi)
+    i16_inc(cursor_pointer_lo)
 
     lda #'x'
     sta (cursor_pointer_lo, x)
-    i16_inc(cursor_pointer_lo, cursor_pointer_hi)
+    i16_inc(cursor_pointer_lo)
 
     lda zp_temp
 
@@ -173,7 +173,7 @@ print_hex:
 
     lda !hex_chars+, y
     sta (cursor_pointer_lo, x)
-    i16_inc(cursor_pointer_lo, cursor_pointer_hi)
+    i16_inc(cursor_pointer_lo)
 
     lda zp_temp
     and #$0f
@@ -181,7 +181,7 @@ print_hex:
 
     lda !hex_chars+, y
     sta (cursor_pointer_lo, x)
-    i16_inc(cursor_pointer_lo, cursor_pointer_hi)
+    i16_inc(cursor_pointer_lo)
 
     rts
 
@@ -205,7 +205,7 @@ print_decimal_u16:
     lda $25
     sta $23
 
-    i16_mul10($22, $23)
+    i16_mul10($22)
 
     i16_i16_sub($22, $23, $03, $04, $22, $23)