ref: 49a810aeb5cb996bf9a6d4e3c38fc4d17fb77610
dir: /d_scan.c/
#include "quakedef.h" #include "r_fog.h" int *r_turb_turb; /* ============= D_WarpScreen // this performs a slight compression of the screen at the same time as // the sine warp, to keep the edges from wrapping ============= */ void D_WarpScreen (void) { static pixel_t *rowptr[MAXHEIGHT+(AMP2*2)]; static int column[MAXWIDTH+(AMP2*2)]; int w, h, u, v, *turb, *col; pixel_t *dest, **row; float wratio, hratio; w = r_refdef.vrect.width; h = r_refdef.vrect.height; wratio = w / (float)scr_vrect.width; hratio = h / (float)scr_vrect.height; for(v = 0; v < scr_vrect.height+AMP2*2; v++) rowptr[v] = dvars.viewbuffer + (r_refdef.vrect.y * dvars.width) + (dvars.width * (int)((float)v * hratio * h / (h + AMP2 * 2))); for(u = 0; u < scr_vrect.width+AMP2*2; u++) column[u] = r_refdef.vrect.x + (int)((float)u * wratio * w / (w + AMP2 * 2)); turb = intsintable + ((int)(cl.time*SPEED)&(CYCLE-1)); dest = vid.buffer + scr_vrect.y * vid.width + scr_vrect.x; for(v = 0; v < scr_vrect.height; v++, dest += vid.width){ col = &column[turb[v]]; row = &rowptr[v]; for(u = 0; u < scr_vrect.width; u += 4){ dest[u+0] = row[turb[u+0]][col[u+0]]; dest[u+1] = row[turb[u+1]][col[u+1]]; dest[u+2] = row[turb[u+2]][col[u+2]]; dest[u+3] = row[turb[u+3]][col[u+3]]; } } } static inline void dospan_solid(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, int width, uzint *pz, uzint izi, int izistep) { pixel_t pix; do{ *pz++ = izi; pix = pbase[(s >> 16) + (t >> 16) * width]; s += sstep; t += tstep; izi += izistep; *pdest++ = pix; }while(--spancount); } static inline void dospan_solid_f1(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, int width, uzint *pz, uzint izi, int izistep, fog_t *fog) { pixel_t pix; do{ *pz++ = izi; izi += izistep; pix = pbase[(s >> 16) + (t >> 16) * width]; s += sstep; t += tstep; *pdest++ = blendfog(pix, *fog); fogstep(*fog); }while(--spancount); } static inline void dospan_blend(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, int width, byte alpha, uzint *pz, uzint izi, int izistep) { pixel_t pix; do{ pix = pbase[(s >> 16) + (t >> 16) * width]; s += sstep; t += tstep; if(opaque(pix) && *pz <= izi) *pdest = blendalpha(pix, *pdest, alpha); izi += izistep; pdest++; pz++; }while(--spancount); } static inline void dospan_blend_f1(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, int width, byte alpha, uzint *pz, uzint izi, int izistep, fog_t *fog) { pixel_t pix; do{ pix = pbase[(s >> 16) + (t >> 16) * width]; s += sstep; t += tstep; if(opaque(pix) && *pz <= izi) *pdest = blendalpha(blendfog(pix, *fog), *pdest, alpha); izi += izistep; pdest++; pz++; fogstep(*fog); }while(--spancount); } static inline void dospan_fence(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, int width, uzint *pz, uzint izi, int izistep) { pixel_t pix; do{ pix = pbase[(s >> 16) + (t >> 16) * width]; s += sstep; t += tstep; if(opaque(pix) && *pz <= izi){ *pdest = pix; *pz = izi; } izi += izistep; pdest++; pz++; }while(--spancount); } static inline void dospan_fence_f1(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, int width, uzint *pz, uzint izi, int izistep, fog_t *fog) { pixel_t pix; do{ pix = pbase[(s >> 16) + (t >> 16) * width]; s += sstep; t += tstep; if(opaque(pix) && *pz <= izi){ *pdest = blendfog(pix, *fog); *pz = izi; } izi += izistep; pdest++; pz++; fogstep(*fog); }while(--spancount); } static void dospan_turb(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, byte alpha, uzint *pz, uzint izi, int izistep) { int sturb, tturb; bool noblend; noblend = (r_drawflags & DRAW_BLEND) == 0; s &= (CYCLE<<16)-1; t &= (CYCLE<<16)-1; do{ if(noblend || *pz <= izi){ sturb = ((s + r_turb_turb[(t>>16)&(CYCLE-1)])>>16)&63; tturb = ((t + r_turb_turb[(s>>16)&(CYCLE-1)])>>16)&63; *pdest = blendalpha(*(pbase + (tturb<<6) + sturb), *pdest, alpha); *pz = izi; // FIXME(sigrid): can always update this one? } s += sstep; t += tstep; izi += izistep; pdest++; pz++; }while(--spancount > 0); } static void dospan_turb_f1(pixel_t *pdest, pixel_t *pbase, int s, int t, int sstep, int tstep, int spancount, byte alpha, uzint *pz, uzint izi, int izistep, fog_t *fog) { int sturb, tturb; bool noblend; noblend = (r_drawflags & DRAW_BLEND) == 0; s &= (CYCLE<<16)-1; t &= (CYCLE<<16)-1; do{ if(noblend || *pz <= izi){ sturb = ((s + r_turb_turb[(t>>16)&(CYCLE-1)])>>16)&63; tturb = ((t + r_turb_turb[(s>>16)&(CYCLE-1)])>>16)&63; *pdest = blendalpha(blendfog(*(pbase + (tturb<<6) + sturb), *fog), *pdest, alpha); *pz = izi; // FIXME(sigrid): can always update this one? } s += sstep; t += tstep; izi += izistep; pdest++; pz++; fogstep(*fog); }while(--spancount > 0); } void D_DrawSpans16(espan_t *pspan, pixel_t *pbase, int width, byte alpha, int spanfunc) { int count, spancount, izistep, spancountminus1; pixel_t *pdest; uzint *pz, izi; fixed16_t s, t, snext, tnext, sstep, tstep; float sdivz, tdivz, zi, z, du, dv; float sdivzstepu, tdivzstepu, zistepu; fog_t fog; bool fogged; sstep = 0; // keep compiler happy tstep = 0; // ditto memset(&fog, 0, sizeof(fog)); sdivzstepu = dvars.sdivzstepu * 16; tdivzstepu = dvars.tdivzstepu * 16; zistepu = dvars.zistepu * 16; izistep = (int)(dvars.zistepu * 0x8000 * 0x10000); do{ pdest = dvars.viewbuffer + pspan->v*dvars.width + pspan->u; pz = dvars.zbuffer + pspan->v*dvars.width + pspan->u; zi = dvars.ziorigin + pspan->v*dvars.zistepv + pspan->u*dvars.zistepu; izi = zi * 0x8000 * 0x10000; count = pspan->count; // calculate the initial s/z, t/z, 1/z, s, and t and clamp du = pspan->u; dv = pspan->v; sdivz = dvars.sdivzorigin + dv*dvars.sdivzstepv + du*dvars.sdivzstepu; tdivz = dvars.tdivzorigin + dv*dvars.tdivzstepv + du*dvars.tdivzstepu; z = (float)0x10000 / zi; // prescale to 16.16 fixed-point s = (int)(sdivz * z) + dvars.sadjust; s = clamp(s, 0, dvars.bbextents); t = (int)(tdivz * z) + dvars.tadjust; t = clamp(t, 0, dvars.bbextentt); do{ // calculate s and t at the far end of the span spancount = min(count, 16); count -= spancount; if(count){ // calculate s/z, t/z, zi->fixed s and t at far end of span, // calculate s and t steps across span by shifting sdivz += sdivzstepu; tdivz += tdivzstepu; // prescale to 16.16 fixed-point z = (float)0x10000 / (zi + zistepu); snext = (int)(sdivz * z) + dvars.sadjust; // prevent round-off error on <0 steps from // from causing overstepping & running off the // edge of the texture snext = clamp(snext, 16, dvars.bbextents); tnext = (int)(tdivz * z) + dvars.tadjust; // guard against round-off error on <0 steps tnext = clamp(tnext, 16, dvars.bbextentt); sstep = (snext - s) >> 4; tstep = (tnext - t) >> 4; }else{ // calculate s/z, t/z, zi->fixed s and t at last pixel in span (so // can't step off polygon), clamp, calculate s and t steps across // span by division, biasing steps low so we don't run off the // texture spancountminus1 = spancount - 1; sdivz += dvars.sdivzstepu * spancountminus1; tdivz += dvars.tdivzstepu * spancountminus1; // prescale to 16.16 fixed-point z = (float)0x10000 / (zi + dvars.zistepu * spancountminus1); snext = (int)(sdivz * z) + dvars.sadjust; // prevent round-off error on <0 steps from // from causing overstepping & running off the // edge of the texture snext = clamp(snext, 16, dvars.bbextents); tnext = (int)(tdivz * z) + dvars.tadjust; // guard against round-off error on <0 steps tnext = clamp(tnext, 16, dvars.bbextentt); if(spancount > 1){ sstep = (snext - s) / spancountminus1; tstep = (tnext - t) / spancountminus1; } } fogged = isfogged() ? fogcalc(izi, izi + izistep*spancount, spancount, &fog) : false; if(fogged){ switch(spanfunc){ case SPAN_SOLID: dospan_solid_f1(pdest, pbase, s, t, sstep, tstep, spancount, width, pz, izi, izistep, &fog); break; case SPAN_TURB: dospan_turb_f1(pdest, pbase, s, t, sstep, tstep, spancount, alpha, pz, izi, izistep, &fog); break; case SPAN_BLEND: dospan_blend_f1(pdest, pbase, s, t, sstep, tstep, spancount, width, alpha, pz, izi, izistep, &fog); break; case SPAN_FENCE: dospan_fence_f1(pdest, pbase, s, t, sstep, tstep, spancount, width, pz, izi, izistep, &fog); break; } }else{ switch(spanfunc){ case SPAN_SOLID: dospan_solid(pdest, pbase, s, t, sstep, tstep, spancount, width, pz, izi, izistep); break; case SPAN_TURB: dospan_turb(pdest, pbase, s, t, sstep, tstep, spancount, alpha, pz, izi, izistep); break; case SPAN_BLEND: dospan_blend(pdest, pbase, s, t, sstep, tstep, spancount, width, alpha, pz, izi, izistep); break; case SPAN_FENCE: dospan_fence(pdest, pbase, s, t, sstep, tstep, spancount, width, pz, izi, izistep); break; } } pdest += spancount; pz += spancount; izi += izistep*spancount; zi += zistepu; s = snext; t = tnext; }while(count > 0); }while((pspan = pspan->pnext) != nil); }