pp14 = (p12 - p15) * cos1_8;
    pp15 = (p13 - p14) * cos3_8;
    p0 = pp0 + pp1;
    p1 = (pp0 - pp1) * cos1_4;
    p2 = pp2 + pp3;
    p3 = (pp2 - pp3) * cos1_4;
    p4 = pp4 + pp5;
    p5 = (pp4 - pp5) * cos1_4;
    p6 = pp6 + pp7;
    p7 = (pp6 - pp7) * cos1_4;
    p8 = pp8 + pp9;
    p9 = (pp8 - pp9) * cos1_4;
    p10 = pp10 + pp11;
    p11 = (pp10 - pp11) * cos1_4;
    p12 = pp12 + pp13;
    p13 = (pp12 - pp13) * cos1_4;
    p14 = pp14 + pp15;
    p15 = (pp14 - pp15) * cos1_4;
    // this is pretty insane coding
    float tmp1;
    new_v19/*
         *  36-17
         */
    = -(new_v4 = (new_v12 = p7) + p5) - p6;
    new_v27/*
         *  44-17
         */
    = -p6 - p7 - p4;
    new_v6 = (new_v10 = (new_v14 = p15) + p11) + p13;
    new_v17/*
         *  34-17
         */
    = -(new_v2 = p15 + p13 + p9) - p14;
    new_v21/*
         *  38-17
         */
    = (tmp1 = -p14 - p15 - p10 - p11) - p13;
    new_v29/*
         *  46-17
         */
    = -p14 - p15 - p12 - p8;
    new_v25/*
         *  42-17
         */
    = tmp1 - p12;
    new_v31/*
         *  48-17
         */
    = -p0;
    new_v0 = p1;
    new_v23/*
         *  40-17
         */
    = -(new_v8 = p3) - p2;
    p0 = (smp1 - smp32) * cos1_64;
    p1 = (smp2 - smp31) * cos3_64;
    p2 = (smp3 - smp30) * cos5_64;
    p3 = (smp4 - smp29) * cos7_64;
    p4 = (smp5 - smp28) * cos9_64;
    p5 = (smp6 - smp27) * cos11_64;
    p6 = (smp7 - smp26) * cos13_64;
    p7 = (smp8 - smp25) * cos15_64;
    p8 = (smp9 - smp24) * cos17_64;
    p9 = (smp10 - smp23) * cos19_64;
    p10 = (smp11 - smp22) * cos21_64;
    p11 = (smp12 - smp21) * cos23_64;
    p12 = (smp13 - smp20) * cos25_64;
    p13 = (smp14 - smp19) * cos27_64;
    p14 = (smp15 - smp18) * cos29_64;
    p15 = (smp16 - smp17) * cos31_64;
    pp0 = p0 + p15;
    pp1 = p1 + p14;
    pp2 = p2 + p13;
    pp3 = p3 + p12;
    pp4 = p4 + p11;
    pp5 = p5 + p10;
    pp6 = p6 + p9;
    pp7 = p7 + p8;
    pp8 = (p0 - p15) * cos1_32;
    pp9 = (p1 - p14) * cos3_32;
    pp10 = (p2 - p13) * cos5_32;
    pp11 = (p3 - p12) * cos7_32;
    pp12 = (p4 - p11) * cos9_32;
    pp13 = (p5 - p10) * cos11_32;
    pp14 = (p6 - p9) * cos13_32;
    pp15 = (p7 - p8) * cos15_32;
    p0 = pp0 + pp7;
    p1 = pp1 + pp6;
    p2 = pp2 + pp5;
    p3 = pp3 + pp4;
    p4 = (pp0 - pp7) * cos1_16;
    p5 = (pp1 - pp6) * cos3_16;
    p6 = (pp2 - pp5) * cos5_16;
    p7 = (pp3 - pp4) * cos7_16;
    p8 = pp8 + pp15;
    p9 = pp9 + pp14;
    p10 = pp10 + pp13;
    p11 = pp11 + pp12;
    p12 = (pp8 - pp15) * cos1_16;
    p13 = (pp9 - pp14) * cos3_16;
    p14 = (pp10 - pp13) * cos5_16;
    p15 = (pp11 - pp12) * cos7_16;
    pp0 = p0 + p3;
    pp1 = p1 + p2;
    pp2 = (p0 - p3) * cos1_8;
    pp3 = (p1 - p2) * cos3_8;
    pp4 = p4 + p7;
    pp5 = p5 + p6;
    pp6 = (p4 - p7) * cos1_8;
    pp7 = (p5 - p6) * cos3_8;
    pp8 = p8 + p11;
    pp9 = p9 + p10;
    pp10 = (p8 - p11) * cos1_8;
    pp11 = (p9 - p10) * cos3_8;
    pp12 = p12 + p15;
    pp13 = p13 + p14;
    pp14 = (p12 - p15) * cos1_8;
    pp15 = (p13 - p14) * cos3_8;
    p0 = pp0 + pp1;
    p1 = (pp0 - pp1) * cos1_4;
    p2 = pp2 + pp3;
    p3 = (pp2 - pp3) * cos1_4;
    p4 = pp4 + pp5;
    p5 = (pp4 - pp5) * cos1_4;
    p6 = pp6 + pp7;
    p7 = (pp6 - pp7) * cos1_4;
    p8 = pp8 + pp9;
    p9 = (pp8 - pp9) * cos1_4;
    p10 = pp10 + pp11;
    p11 = (pp10 - pp11) * cos1_4;
    p12 = pp12 + pp13;
    p13 = (pp12 - pp13) * cos1_4;
    p14 = pp14 + pp15;
    p15 = (pp14 - pp15) * cos1_4;
    // manually doing something that a compiler should handle sucks
    // coding like this is hard to read
    float tmp2;
    new_v5 = (new_v11 = (new_v13 = (new_v15 = p15) + p7) + p11) + p5 + p13;
    new_v7 = (new_v9 = p15 + p11 + p3) + p13;
    new_v16/*
         *  33-17
         */
    = -(new_v1 = (tmp1 = p13 + p15 + p9) + p1) - p14;
    new_v18/*
         *  35-17
         */
    = -(new_v3 = tmp1 + p5 + p7) - p6 - p14;
    new_v22/*
         *  39-17
         */
    = (tmp1 = -p10 - p11 - p14 - p15) - p13 - p2 - p3;
    new_v20/*
         *  37-17
         */
    = tmp1 - p13 - p5 - p6 - p7;
    new_v24/*
         *  41-17
         */
    = tmp1 - p12 - p2 - p3;
    new_v26/*
         *  43-17
         */
    = tmp1 - p12 - (tmp2 = p4 + p6 + p7);
    new_v30/*
         *  47-17
         */
    = (tmp1 = -p8 - p12 - p14 - p15) - p0;
    new_v28/*
         *  45-17
         */
    = tmp1 - tmp2;
    loc1 = actual_write_pos;
     loc2 = 16 + actual_write_pos;
     loc3 = 32 + actual_write_pos;
     loc4 = 48 + actual_write_pos;
     loc5 = 64 + actual_write_pos;
     loc6 = 80 + actual_write_pos;
     loc7 = 96 + actual_write_pos;
     loc8 = 112 + actual_write_pos;
     loc9 = 128 + actual_write_pos;
     loc10 = 144 + actual_write_pos;
     loc11 = 160 + actual_write_pos;
     loc12 = 176 + actual_write_pos;
     loc13 = 192 + actual_write_pos;
     loc14 = 208 + actual_write_pos;
     loc15 = 224 + actual_write_pos;
     loc16 = 240 + actual_write_pos;
     loc17 = 256 + actual_write_pos;
     loc18 = 272 + actual_write_pos;
     loc19 = 288 + actual_write_pos;
     loc20 = 304 + actual_write_pos;
     loc21 = 320 + actual_write_pos;
     loc22 = 336 + actual_write_pos;
     loc23 = 352 + actual_write_pos;
     loc24 = 368 + actual_write_pos;
     loc25 = 384 + actual_write_pos;
     loc26 = 400 + actual_write_pos;
     loc27 = 416 + actual_write_pos;
     loc28 = 432 + actual_write_pos;
     loc29 = 448 + actual_write_pos;
     loc30 = 464 + actual_write_pos;
     loc31 = 480 + actual_write_pos;
     loc32 = 496 + actual_write_pos;
    float dest[] = actual_v;
    dest[loc1] = new_v0;
    dest[loc2] = new_v1;
    dest[loc3] = new_v2;
    dest[loc4] = new_v3;
    dest[loc5] = new_v4;
    dest[loc6] = new_v5;
    dest[loc7] = new_v6;
    dest[loc8] = new_v7;
    dest[loc9] = new_v8;
    dest[loc10] = new_v9;
    dest[loc11] = new_v10;
    dest[loc12] = new_v11;
    dest[loc13] = new_v12;
    dest[loc14] = new_v13;
    dest[loc15] = new_v14;
    dest[loc16] = new_v15;
    dest[loc17] = 0.0f;
    dest[loc18] = -new_v15;
    dest[loc19] = -new_v14;
    dest[loc20] = -new_v13;
    dest[loc21] = -new_v12;
    dest[loc22] = -new_v11;
    dest[loc23] = -new_v10;
    dest[loc24] = -new_v9;
    dest[loc25] = -new_v8;
    dest[loc26] = -new_v7;
    dest[loc27] = -new_v6;
    dest[loc28] = -new_v5;
    dest[loc29] = -new_v4;
    dest[loc30] = -new_v3;
    dest[loc31] = -new_v2;
    dest[loc32] = -new_v1;
    dest = (dest == v1) ? v2 : v1;
    dest[loc1] = -new_v0;
    dest[loc2] = new_v16;
    dest[loc3] = new_v17;
    dest[loc4] = new_v18;
    dest[loc5] = new_v19;
    dest[loc6] = new_v20;
    dest[loc7] = new_v21;
    dest[loc8] = new_v22;
    dest[loc9] = new_v23;
    dest[loc10] = new_v24;
    dest[loc11] = new_v25;
    dest[loc12] = new_v26;
    dest[loc13] = new_v27;