Prototyping DOS effects with ShaderToy - Revision history

Pestis: /* Vector arithmetic (examples) */

2025-10-09T08:57:55Z

‎Vector arithmetic (examples)

2025-10-09T08:48:09Z

2025-10-09T08:19:31Z

2025-10-09T08:12:40Z

‎Scalar functions

2025-10-09T07:56:57Z

‎Rounding and remainders

2025-10-07T17:51:45Z

2025-10-07T17:50:35Z

2025-10-07T17:46:34Z

2025-10-07T17:42:48Z

2025-10-07T09:53:28Z

@@ Line 101: / Line 101: @@
 |-
 || length(a) || 14 || If a is not needed later: fmul st0, st0; fxch st0, st2; fmul st0, st0; faddp st2, st0; fmul st0, st0; faddp st1, st0; fsqrt
 |}

@@ Line 43: / Line 43: @@
 |-
 | log2(x) || 4 || fld1<br>...<br>fyl2x
 |-
 | mix(x,y,a) || 10 || Stack in: a x y. Stack out: x*(1-a)+y*a.<br/>fmul st2, st0; fld1; fsubrp st1, st0; fmulp st1, st0; faddp st1, st0

@@ Line 43: / Line 43: @@
 |-
 | log2(x) || 4 || fld1<br>...<br>fyl2x
 |-
 | 2*min(x,y) || 10 || Computed as a+b-abs(a-b) i.e.<br/>fld st0; fsub st0, st2; fabs; fsubp st1, st0; faddp st1, st0<br/>Replace fsubp with faddp for 2*max(x,y)

@@ Line 49: / Line 49: @@
 |-
 | exp2(y) || 14 || fld1; fld st1; fprem; f2xm1; faddp st1,st0; fscale; fstp st1
 |-
 | pow(x,y) || 16 || Computed as 2^(y*log2(x)) i.e. fyl2x, followed by the exp2(x) code

@@ Line 69: / Line 69: @@
 | mod(x,y) || 2 || fprem or fprem1. Notice they compute the remainder, not modulo, so they are the same only for positive values of x.
 |-
-| ceil(x) || 2-7 || Up to 5 bytes to setup the rounding mode with fldcw, followed by frndint
+| ceil(x) || 2-7 || Up to 5 bytes to setup the rounding mode with fldcw (RC field = 10, round up), followed by frndint
 |-
-| floor(x) || 2-7 || Up to 5 bytes to setup the rounding mode with fldcw, followed by frndint
+| floor(x) || 2-7 || Up to 5 bytes to setup the rounding mode with fldcw (RC field = 01, round down), followed by frndint
 |}

@@ Line 144: / Line 144: @@
 <pre>
      mov     cl, ITERS
-.maploop:
+.maploop:                ; for(int j=0;j<ITERS;j++) {
      fld     st0
      frndint
@@ Line 163: / Line 163: @@
      fmul    dword [si]
      fsubp   st3, st0     ; t.z -= t.x * o
-     loop    .maploop
+     loop    .maploop     ; }
 </pre>

@@ Line 145: / Line 145: @@
      mov     cl, ITERS
 .maploop:
-     fld     st0          ; t.x t.x
+     fld     st0
      frndint
-     fsubp   st1, st0     ; t.x-round(t.x)
+     fsubp   st1, st0
      fabs                 ; t.x = abs(t.x - round(t.x))
      fadd    st0          ; t.x += t.x;

@@ Line 95: / Line 95: @@
 |}
-From this you can already see that a simple <code>normalize(x)</code> is going to take a lot of bytes, as it has to be computed as <code>x/=length(x)</code>. Therefore, normalizing your raymarchers rays is usually to be avoided. <code>cross</code>, <code>reflect</code>, and <code>refract</code> are probably also too costly for sizecoding.
+From this you can already see that a simple <code>normalize(a)</code> is going to take a lot of bytes, as it has to be computed as <code>a/=length(a)</code>. Therefore, normalizing your raymarchers rays is usually to be avoided. <code>cross</code>, <code>reflect</code>, and <code>refract</code> are probably also too costly for sizecoding.
 == Floating point constants ==

@@ Line 92: / Line 92: @@
 || a+=b || 10 || If b is needed later: fadd st3; fld st1; faddp st5; fld st2; faddp st6
 |-
-|| length(a) || 16 || If a is not needed later: fmul st0, st0; fld st1; fmul st0, st0; faddp st1, st0; fld st2; fmul st0, st0; faddp st1, st0; fsqrt;
+|| length(a) || 14 || If a is not needed later: fmul st0, st0; fxch st0, st2; fmul st0, st0; faddp st2, st0; fmul st0, st0; faddp st1, st0; fsqrt
 |}

@@ Line 44: / Line 44: @@
 | log2(x) || 4 || fld1<br>...<br>fyl2x
 |-
-| 2*min(x,y) || 10 || Computed as a+b-abs(a-b) i.e.<br/>fld st0; fsub st0, st2; fabs; fsubp st1, st0; faddp st1, st0<br/>Replace fsubp with faddp for max(x,y)
+| 2*min(x,y) || 10 || Computed as a+b-abs(a-b) i.e.<br/>fld st0; fsub st0, st2; fabs; fsubp st1, st0; faddp st1, st0<br/>Replace fsubp with faddp for 2*max(x,y)
 |-
-| min(x,y) || 11 || fcom st0, st1; fnstsw ax; sahf; jc .S; fxch st0, st1; .S: fstp st1, st0<br/>Replace jc with jnc for 2*max(x,y)
+| min(x,y) || 11 || fcom st0, st1; fnstsw ax; sahf; jc .S; fxch st0, st1; .S: fstp st1, st0<br/>Replace jc with jnc for max(x,y)
 |-
 | exp2(y) || 14 || fld1; fld st1; fprem; f2xm1; faddp st1,st0; fscale; fstp st1