Noise wave generation code optimized.

README.md

@@ -1,7 +1,7 @@
 # Raspberry Casket
 A fast and small open source Pretracker replayer
 
-## Raspberry Casket Player V2.x (10-Aug-2023)
+## Raspberry Casket Player V2.x (22-Aug-2023)
 
 Provided by Chris 'platon42' Hodges <chrisly@platon42.de>
 
@@ -101,7 +101,7 @@ The original code compressed with *Blueberry's* Shrinkler goes from
 18052 bytes down to 9023 bytes.
 
 Raspberry Casket, depending on the features compiled in, is about
-5784 bytes and shrinkles down to ~4118 bytes (in isolation).
+5732 bytes and shrinkles down to ~4084 bytes (in isolation).
 
 So this means that the optimization is not just "on the outside".
 
@@ -168,8 +168,9 @@ solve this problem.
 - Bugfix: Songend detection for back-jumps was broken since at least V1.1.
 - Optimized some more wave selection code.
 - Nosync/sync wave selection optimized.
+- Optimized wave generation a bit more (noise generator).
 - Added Presto player draft.
-- Drop-in replacement code size: 5784 bytes.
+- Drop-in replacement code size: 5732 bytes.
 
 ### V1.x (unreleased)
 - Fixed a bug regarding the copper output mode with looping waves having a loop-offset.

src/raspberry_casket.asm

@@ -1,5 +1,5 @@
 ;--------------------------------------------------------------------
-; Raspberry Casket Player V2.x (20-Aug-2023)
+; Raspberry Casket Player V2.x (22-Aug-2023)
 ; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 ;
 ; Provided by Chris 'platon42' Hodges <chrisly@platon42.de>
@@ -96,7 +96,7 @@
 ; 18052 bytes down to 9023 bytes.
 ;
 ; Raspberry Casket, depending on the features compiled in, is about
-; 5784 bytes and shrinkles down to ~4118 bytes (in isolation).
+; 5732 bytes and shrinkles down to ~4084 bytes (in isolation).
 ;
 ; So this means that the optimization is not just "on the outside".
 ;

src/raspberry_casket_wavegen.asm

@@ -1,25 +1,25 @@
 
 CLIPTO8BIT MACRO
-        cmpi.w  #-$80,\1
+        cmp.w   #-$80,\1
         bge.s   .nominclip\@
         moveq.l #-$80,\1
 .nominclip\@
-        cmpi.w  #$7F,\1
+        cmp.w   #$7f,\1
         ble.s   .nomaxclip\@
-        moveq.l #$7F,\1
+        moveq.l #$7f,\1
 .nomaxclip\@
         ENDM
 
 CLIPORTRUNC8BIT MACRO
         beq.s   .unboosted\@
-        asr.l    #6,\1
-        cmpi.w  #-$80,\1
+        asr.l   #6,\1
+        cmp.w   #-$80,\1
         bge.s   .nominclip\@
         moveq.l #-$80,\1
 .nominclip\@
-        cmpi.w  #$7F,\1
+        cmp.w   #$7f,\1
         ble.s   .nomaxclip\@
-        moveq.l #$7F,\1
+        moveq.l #$7f,\1
         bra.s   .nomaxclip\@
 .unboosted\@
         asr.l   #8,\1
@@ -254,8 +254,10 @@ pre_WaveGen:
         bne     .no_noise
 
 ; ----------------------------------------
+; d0 = scratch
 ; d1 = octave
 ; d2 = pitch ramping value
+; d4 = scratch
 ; a2 = base note
 .gen_noise
         IFNE    PRETRACKER_PARANOIA_MODE
@@ -307,6 +309,8 @@ pre_WaveGen:
         move.b  wi_chord_shift_b(a3),d1
         add.w   d1,d0
         addq.w  #1,d0
+        
+        ; sum of phase min and shift are used as root for noise
 
         movea.l d5,a1
         movea.l pv_wg_curr_sample_ptr(a4),a0
@@ -330,30 +334,26 @@ pre_WaveGen:
         lsl.w   #7,d1
         eor.w   d1,d0
 
-        cmpa.l  #$8000,a5       ; if symmetrical
-        beq.s   .gen_noise_centered
-        ; FIXME what does this do? d4 = (a5 - $8000) d1 = (a5 + $7fff)&$ffff8000 -> d4 - d1 == ((a5 - $8000) - ((a5 + $7fff)&$ffff8000)
-        move.l  a5,d4
-        addi.l  #$FFFF8000,d4
-        move.l  a5,d1
-        addi.l  #$FFFF7FFF,d1
-        andi.w  #$8000,d1
-        movea.l d4,a5
-        suba.l  d1,a5
-.gen_noise_centered
-
-        move.b  d0,d1
+        move.b  d0,d1           ; take the random seed
         ext.w   d1
-        move.w  d3,d4
-        muls    d1,d4
-        move.l  d4,d1
-        asr.l   #7,d1
+        muls    d3,d1           ; multiply by gain
+        asr.w   #7,d1
         CLIPTO8BIT d1
         add.b   (a0),d1
 
+        move.w  a5,d4
+        subq.w  #1,d4
+        and.w   #$7fff,d4
+        move.w  d4,a5
+        addq.w  #1,a5
+
 .gen_noise_innerloop
         move.b  d1,(a0)+
+        cmpa.l  pv_wg_curr_samend_ptr(a4),a0
+        beq     .wave_gen_tone_done
+
         adda.l  a1,a5
+
         tst.l   d2
         beq.s   .gen_noise_no_pitch_ramping
         add.l   d2,d6
@@ -365,22 +365,21 @@ pre_WaveGen:
 
         btst    #4,wi_flags_b(a3)   ; pitch linear flag
         beq.s   .noise_nonlinear_pitch
-        move.l  d2,d4
+        move.l  d2,d4               ; filter pitch speed
         asr.l   #7,d4
         sub.l   d4,d2
 .noise_nonlinear_pitch
 
-        cmpa.w  #$1FF,a1
-        bgt.s   .gen_noise_no_pitch_ramping
-        moveq.l #0,d2
+        cmpa.w  #$1ff,a1
+        bgt.s   .gen_noise_no_end_of_pitch_ramp
+        moveq.l #0,d2               ; stop pitch ramping
         move.l  d2,a5
-        ;suba.l   a5,a5
         movea.w #$200,a1
+.gen_noise_no_end_of_pitch_ramp
 .gen_noise_no_pitch_ramping
-        cmpa.l  pv_wg_curr_samend_ptr(a4),a0
-        beq     .wave_gen_tone_done
-        cmpa.w  #$7FFF,a5
-        ble.s   .gen_noise_innerloop
+        cmpa.w  a5,a5
+        beq.s   .gen_noise_innerloop
+
         bra     .gen_noise_outerloop
 
 ; ----------------------------------------