archive/blender-archive
Archived
1
1
Fork 0
Code Pull Requests Packages Activity

nurbs surface resolution is now per segment rather then being distributed over the whole surface.

Browse Source 
This is what 2D do alredy and makes resolution settings more like subsurf level.

- This makes it impossible to have a lower resolution output then the nurbs cage, however this dosnt seem a very useful feature.
- Using the do-versions from recent curve interpolation fix.

Tested with these files.

http://digilander.libero.it/pafurijaz/Files/Blender/Libellula.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Martello.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Punta.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Phones.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Tubo.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Pipe.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Tire.zip
http://digilander.libero.it/pafurijaz/Files/Blender/Cotton.zip
This commit is contained in:
Campbell Barton
2008-09-26 19:00:49 +00:00
parent 8e1cf42dbd
commit 3ad55e1541
4 changed files with 46 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
source/blender/blenkernel/intern/curve.c
View File

@@ -658,23 +658,23 @@ void makeNurbfaces(Nurb *nu, float *coord_array, int rowstride)
float u, v, ustart, uend, ustep, vstart, vend, vstep, sumdiv;
int i, j, iofs, jofs, cycl, len, resolu, resolv;
int istart, iend, jsta, jen, *jstart, *jend, ratcomp;
int totu = nu->pntsu*nu->resolu, totv = nu->pntsv*nu->resolv;
if(nu->knotsu==NULL || nu->knotsv==NULL) return;
if(nu->orderu>nu->pntsu) return;
if(nu->orderv>nu->pntsv) return;
if(coord_array==NULL) return;
/* allocate and initialize */
len= nu->pntsu*nu->pntsv;
len = totu * totv;
if(len==0) return;
sum= (float *)MEM_callocN(sizeof(float)*len, "makeNurbfaces1");
resolu= nu->resolu;
resolv= nu->resolv;
len= resolu*resolv;
len= totu*totv;
if(len==0) {
MEM_freeN(sum);
return;
@@ -690,12 +690,13 @@ void makeNurbfaces(Nurb *nu, float *coord_array, int rowstride)
}
bp++;
}
fp= nu->knotsu;
ustart= fp[nu->orderu-1];
if(nu->flagu & CU_CYCLIC) uend= fp[nu->pntsu+nu->orderu-1];
else uend= fp[nu->pntsu];
ustep= (uend-ustart)/(resolu-1+(nu->flagu & CU_CYCLIC));
ustep= (uend-ustart)/((nu->flagu & CU_CYCLIC) ? totu : totu - 1);
basisu= (float *)MEM_mallocN(sizeof(float)*KNOTSU(nu), "makeNurbfaces3");
fp= nu->knotsv;
@@ -703,17 +704,19 @@ void makeNurbfaces(Nurb *nu, float *coord_array, int rowstride)
if(nu->flagv & CU_CYCLIC) vend= fp[nu->pntsv+nu->orderv-1];
else vend= fp[nu->pntsv];
vstep= (vend-vstart)/(resolv-1+(nu->flagv & CU_CYCLIC));
vstep= (vend-vstart)/((nu->flagv & CU_CYCLIC) ? totv : totv - 1);
len= KNOTSV(nu);
basisv= (float *)MEM_mallocN(sizeof(float)*len*resolv, "makeNurbfaces3");
jstart= (int *)MEM_mallocN(sizeof(float)*resolv, "makeNurbfaces4");
jend= (int *)MEM_mallocN(sizeof(float)*resolv, "makeNurbfaces5");
basisv= (float *)MEM_mallocN(sizeof(float)*len*totv, "makeNurbfaces3");
jstart= (int *)MEM_mallocN(sizeof(float)*totv, "makeNurbfaces4");
jend= (int *)MEM_mallocN(sizeof(float)*totv, "makeNurbfaces5");
/* precalculation of basisv and jstart,jend */
if(nu->flagv & CU_CYCLIC) cycl= nu->orderv-1;
else cycl= 0;
v= vstart;
basis= basisv;
resolv= totv;
while(resolv--) {
basisNurb(v, nu->orderv, (short)(nu->pntsv+cycl), nu->knotsv, basis, jstart+resolv, jend+resolv);
basis+= KNOTSV(nu);
@@ -724,12 +727,13 @@ void makeNurbfaces(Nurb *nu, float *coord_array, int rowstride)
else cycl= 0;
in= coord_array;
u= ustart;
resolu= totu;
while(resolu--) {
basisNurb(u, nu->orderu, (short)(nu->pntsu+cycl), nu->knotsu, basisu, &istart, &iend);
basis= basisv;
resolv= nu->resolv;
resolv= totv;
while(resolv--) {
jsta= jstart[resolv];
@@ -798,7 +802,7 @@ void makeNurbfaces(Nurb *nu, float *coord_array, int rowstride)
basis+= KNOTSV(nu);
}
u+= ustep;
if (rowstride!=0) in = (float*) (((unsigned char*) in) + (rowstride - 3*nu->resolv*sizeof(*in)));
if (rowstride!=0) in = (float*) (((unsigned char*) in) + (rowstride - 3*totv*sizeof(*in)));
}
/* free */
@@ -954,8 +958,8 @@ float *make_orco_surf(Object *ob)
See also convertblender.c: init_render_surf()
*/
sizeu = nu->resolu;
sizev = nu->resolv;
sizeu = nu->pntsu*nu->resolu;
sizev = nu->pntsv*nu->resolv;
if (nu->flagu & CU_CYCLIC) sizeu++;
if (nu->flagv & CU_CYCLIC) sizev++;
if(nu->pntsv>1) tot+= sizeu * sizev;
@@ -968,8 +972,8 @@ float *make_orco_surf(Object *ob)
nu= cu->nurb.first;
while(nu) {
if(nu->pntsv>1) {
sizeu = nu->resolu;
sizev = nu->resolv;
sizeu = nu->pntsu*nu->resolu;
sizev = nu->pntsv*nu->resolv;
if (nu->flagu & CU_CYCLIC) sizeu++;
if (nu->flagv & CU_CYCLIC) sizev++;
@@ -990,7 +994,7 @@ float *make_orco_surf(Object *ob)
}
}
else {
float *_tdata= MEM_callocN(nu->resolu*nu->resolv*3*sizeof(float), "temp data");
float *_tdata= MEM_callocN((nu->pntsu*nu->resolu) * (nu->pntsv*nu->resolv) *3*sizeof(float), "temp data");
float *tdata= _tdata;
makeNurbfaces(nu, tdata, 0);
@@ -1005,7 +1009,7 @@ float *make_orco_surf(Object *ob)
if (a==sizev-1 && (nu->flagv & CU_CYCLIC))
use_a= 0;
tdata = _tdata + 3 * (use_b * nu->resolv + use_a);
tdata = _tdata + 3 * (use_b * (nu->pntsv*nu->resolv) + use_a);
fp[0]= (tdata[0]-cu->loc[0])/cu->size[0];
fp[1]= (tdata[1]-cu->loc[1])/cu->size[1];