diff --git a/polymer/eduke32/build/include/a.h b/polymer/eduke32/build/include/a.h
index 0a8af5826..9f7411c59 100644
--- a/polymer/eduke32/build/include/a.h
+++ b/polymer/eduke32/build/include/a.h
@@ -35,9 +35,9 @@ extern int32_t __cdecl setuptvlineasm2(int32_t,int32_t,int32_t);
 extern int32_t __cdecl tvlineasm2(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t __cdecl mvlineasm1(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t __cdecl setupvlineasm(int32_t);
-extern int32_t __cdecl vlineasm4(int32_t,int32_t);
+extern int32_t __cdecl vlineasm4(int32_t,char *);
 extern int32_t __cdecl setupmvlineasm(int32_t);
-extern int32_t __cdecl mvlineasm4(int32_t,int32_t);
+extern int32_t __cdecl mvlineasm4(int32_t,char *);
 extern int32_t __cdecl setupspritevline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t __cdecl spritevline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t __cdecl msetupspritevline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
@@ -81,9 +81,9 @@ extern int32_t _cdecl setuptvlineasm2(int32_t,int32_t,int32_t);
 extern int32_t _cdecl tvlineasm2(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t _cdecl mvlineasm1(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t _cdecl setupvlineasm(int32_t);
-extern int32_t _cdecl vlineasm4(int32_t,int32_t);
+extern int32_t _cdecl vlineasm4(int32_t,char *);
 extern int32_t _cdecl setupmvlineasm(int32_t);
-extern int32_t _cdecl mvlineasm4(int32_t,int32_t);
+extern int32_t _cdecl mvlineasm4(int32_t,char *);
 extern int32_t _cdecl setupspritevline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t _cdecl spritevline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
 extern int32_t _cdecl msetupspritevline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t);
@@ -109,6 +109,8 @@ extern void _cdecl stretchhline(int32_t,int32_t,int32_t,int32_t,int32_t,int32_t)
 #define ENGINE_USING_A_C
 #include <stdint.h>
 
+#define prevlineasm1 vlineasm1
+
 void setvlinebpl(int32_t dabpl);
 void fixtransluscence(intptr_t datransoff);
 void settransnormal(void);
@@ -123,13 +125,15 @@ void setupslopevlin(int32_t logylogx, intptr_t bufplc, int32_t pinc);
 void slopevlin(intptr_t p, int32_t i, intptr_t slopaloffs, int32_t cnt, int32_t bx, int32_t by);
 
 void setupvlineasm(int32_t neglogy);
-void vlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p);
+int32_t vlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p);
+void vlineasm4(int32_t cnt, char *p);
 
 void setupmvlineasm(int32_t neglogy);
-void mvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p);
+int32_t mvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p);
+void mvlineasm4(int32_t cnt, char *p);
 
 void setuptvlineasm(int32_t neglogy);
-void tvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p);
+int32_t tvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p);
 
 void msethlineshift(int32_t logx, int32_t logy);
 void mhline(intptr_t bufplc, uint32_t bx, int32_t cntup16, int32_t junk, uint32_t by, intptr_t p);
diff --git a/polymer/eduke32/build/src/a-c.c b/polymer/eduke32/build/src/a-c.c
index 4b70b2947..3c7af40e5 100644
--- a/polymer/eduke32/build/src/a-c.c
+++ b/polymer/eduke32/build/src/a-c.c
@@ -89,7 +89,7 @@ void slopevlin(intptr_t p, int32_t i, intptr_t slopaloffs, int32_t cnt, int32_t
 
 void setupvlineasm(int32_t neglogy) { glogy = neglogy; }
 // cnt+1 loop iterations!
-void vlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p)
+int32_t vlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p)
 {
     const char *const buf = (char *)bufplc;
     const char *const pal = (char *)paloffs;
@@ -105,11 +105,58 @@ void vlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intpt
         vplc += vinc;
     }
     while (--cnt);
+
+    return vplc;
+}
+
+
+extern intptr_t palookupoffse[4];
+extern int32_t vplce[4], vince[4];
+extern intptr_t bufplce[4];
+
+// cnt >= 1
+void vlineasm4(int32_t cnt, char *p)
+{
+    char ch;
+    int32_t i;
+#if 1
+    // this gives slightly more stuff in registers in the loop
+    // (on x86_64 at least)
+    char *const pal[4] = {(char *)palookupoffse[0], (char *)palookupoffse[1], (char *)palookupoffse[2], (char *)palookupoffse[3]};
+    char *const buf[4] = {(char *)bufplce[0], (char *)bufplce[1], (char *)bufplce[2], (char *)bufplce[3]};
+    const int32_t vinc[4] = {vince[0], vince[1], vince[2], vince[3]};
+    uint32_t vplc[4] = {vplce[0], vplce[1], vplce[2], vplce[3]};
+#else
+    char *pal[4];
+    char *buf[4];
+    int32_t vinc[4];
+    uint32_t vplc[4];
+
+    Bmemcpy(pal, palookupoffse, sizeof(pal));
+    Bmemcpy(buf, bufplce, sizeof(buf));
+    Bmemcpy(vinc, vince, sizeof(vinc));
+    Bmemcpy(vplc, vplce, sizeof(vplc));
+#endif
+
+    const int32_t logy = glogy, ourbpl = bpl;
+
+    do
+    {
+        for (i=0; i<4; i++)
+        {
+            ch = buf[i][vplc[i]>>logy]; p[i] = pal[i][ch];
+            vplc[i] += vinc[i];
+        }
+        p += ourbpl;
+    }
+    while (--cnt);
+
+    Bmemcpy(vplce, vplc, sizeof(vplce));
 }
 
 void setupmvlineasm(int32_t neglogy) { glogy = neglogy; }
 // cnt+1 loop iterations!
-void mvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p)
+int32_t mvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p)
 {
     char ch;
 
@@ -127,8 +174,38 @@ void mvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intp
         vplc += vinc;
     }
     while (--cnt);
+
+    return vplc;
 }
 
+// cnt >= 1
+void mvlineasm4(int32_t cnt, char *p)
+{
+    char ch;
+    int32_t i;
+
+    char *const pal[4] = {(char *)palookupoffse[0], (char *)palookupoffse[1], (char *)palookupoffse[2], (char *)palookupoffse[3]};
+    char *const buf[4] = {(char *)bufplce[0], (char *)bufplce[1], (char *)bufplce[2], (char *)bufplce[3]};
+    const int32_t vinc[4] = {vince[0], vince[1], vince[2], vince[3]};
+    uint32_t vplc[4] = {vplce[0], vplce[1], vplce[2], vplce[3]};
+
+    const int32_t logy = glogy, ourbpl = bpl;
+
+    do
+    {
+        for (i=0; i<4; i++)
+        {
+            ch = buf[i][vplc[i]>>logy]; if (ch != 255) p[i] = pal[i][ch];
+            vplc[i] += vinc[i];
+        }
+        p += ourbpl;
+    }
+    while (--cnt);
+
+    Bmemcpy(vplce, vplc, sizeof(vplce));
+}
+
+
 void setuptvlineasm(int32_t neglogy) { glogy = neglogy; }
 // cnt+1 loop iterations!
 void tvlineasm1(int32_t vinc, intptr_t paloffs, int32_t cnt, uint32_t vplc, intptr_t bufplc, intptr_t p)
diff --git a/polymer/eduke32/build/src/engine.c b/polymer/eduke32/build/src/engine.c
index f0b6b82db..fe8a84d9e 100644
--- a/polymer/eduke32/build/src/engine.c
+++ b/polymer/eduke32/build/src/engine.c
@@ -2240,7 +2240,8 @@ int32_t xyaspect;
 static int32_t viewingrangerecip;
 
 intptr_t asm1, asm2, asm3, asm4, palookupoffse[4];
-int32_t vplce[4], vince[4], bufplce[4];
+int32_t vplce[4], vince[4];
+intptr_t bufplce[4];
 static char globalxshift, globalyshift;
 static int32_t globalxpanning, globalypanning;
 int32_t globalshade, globalorientation;
@@ -2637,9 +2638,9 @@ skipitaddwall:
 static void maskwallscan(int32_t x1, int32_t x2, int16_t *uwal, int16_t *dwal, int32_t *swal, int32_t *lwal)
 {
     int32_t x,/* startx,*/ xnice, ynice;
-    intptr_t startx, p, fpalookup;
-    int32_t y1ve[4], y2ve[4], /* p,*/ tsizx, tsizy;
-#ifndef ENGINE_USING_A_C
+    intptr_t startx, p, pp, fpalookup;
+    int32_t y1ve[4], y2ve[4], tsizx, tsizy;
+#if 1 //ndef ENGINE_USING_A_C
     char bad;
     int32_t i, u4, d4, dax, z;
 #endif
@@ -2667,7 +2668,7 @@ static void maskwallscan(int32_t x1, int32_t x2, int16_t *uwal, int16_t *dwal, i
 
     setupmvlineasm(globalshiftval);
 
-#ifndef ENGINE_USING_A_C
+#if 1 //ndef ENGINE_USING_A_C
 
     x = startx;
     while ((startumost[x+windowx1] > startdmost[x+windowx1]) && (x <= x2)) x++;
@@ -2741,13 +2742,13 @@ static void maskwallscan(int32_t x1, int32_t x2, int16_t *uwal, int16_t *dwal, i
         if (u4 > y1ve[2]) vplce[2] = mvlineasm1(vince[2],palookupoffse[2],u4-y1ve[2]-1,vplce[2],bufplce[2],ylookup[y1ve[2]]+p+2);
         if (u4 > y1ve[3]) vplce[3] = mvlineasm1(vince[3],palookupoffse[3],u4-y1ve[3]-1,vplce[3],bufplce[3],ylookup[y1ve[3]]+p+3);
 
-        if (d4 >= u4) mvlineasm4(d4-u4+1,ylookup[u4]+p);
+        if (d4 >= u4) mvlineasm4(d4-u4+1, (char *)ylookup[u4]+p);
 
-        i = p+ylookup[d4+1];
-        if (y2ve[0] > d4) mvlineasm1(vince[0],palookupoffse[0],y2ve[0]-d4-1,vplce[0],bufplce[0],i+0);
-        if (y2ve[1] > d4) mvlineasm1(vince[1],palookupoffse[1],y2ve[1]-d4-1,vplce[1],bufplce[1],i+1);
-        if (y2ve[2] > d4) mvlineasm1(vince[2],palookupoffse[2],y2ve[2]-d4-1,vplce[2],bufplce[2],i+2);
-        if (y2ve[3] > d4) mvlineasm1(vince[3],palookupoffse[3],y2ve[3]-d4-1,vplce[3],bufplce[3],i+3);
+        pp = p+ylookup[d4+1];
+        if (y2ve[0] > d4) mvlineasm1(vince[0],palookupoffse[0],y2ve[0]-d4-1,vplce[0],bufplce[0],pp+0);
+        if (y2ve[1] > d4) mvlineasm1(vince[1],palookupoffse[1],y2ve[1]-d4-1,vplce[1],bufplce[1],pp+1);
+        if (y2ve[2] > d4) mvlineasm1(vince[2],palookupoffse[2],y2ve[2]-d4-1,vplce[2],bufplce[2],pp+2);
+        if (y2ve[3] > d4) mvlineasm1(vince[3],palookupoffse[3],y2ve[3]-d4-1,vplce[3],bufplce[3],pp+3);
     }
     for (; x<=x2; x++,p++)
     {
@@ -3686,9 +3687,10 @@ static void wallscan(int32_t x1, int32_t x2, int16_t *uwal, int16_t *dwal, int32
     int32_t x, xnice, ynice;
     intptr_t fpalookup;
     int32_t y1ve[4], y2ve[4], tsizx, tsizy;
-#ifndef ENGINE_USING_A_C
+#if 1 //ndef ENGINE_USING_A_C
     char bad;
     int32_t i, u4, d4, z;
+    uintptr_t p;
 #endif
 #ifdef YAX_ENABLE
     if (g_nodraw)
@@ -3714,7 +3716,7 @@ static void wallscan(int32_t x1, int32_t x2, int16_t *uwal, int16_t *dwal, int32
 
     setupvlineasm(globalshiftval);
 
-#ifndef ENGINE_USING_A_C
+#if 1 //ndef ENGINE_USING_A_C
 
     x = x1;
     while ((umost[x] > dmost[x]) && (x <= x2)) x++;
@@ -3786,13 +3788,13 @@ static void wallscan(int32_t x1, int32_t x2, int16_t *uwal, int16_t *dwal, int32
         if (u4 > y1ve[2]) vplce[2] = prevlineasm1(vince[2],palookupoffse[2],u4-y1ve[2]-1,vplce[2],bufplce[2],ylookup[y1ve[2]]+x+frameoffset+2);
         if (u4 > y1ve[3]) vplce[3] = prevlineasm1(vince[3],palookupoffse[3],u4-y1ve[3]-1,vplce[3],bufplce[3],ylookup[y1ve[3]]+x+frameoffset+3);
 
-        if (d4 >= u4) vlineasm4(d4-u4+1,ylookup[u4]+x+frameoffset);
+        if (d4 >= u4) vlineasm4(d4-u4+1, (char *)ylookup[u4]+x+frameoffset);
 
-        i = x+frameoffset+ylookup[d4+1];
-        if (y2ve[0] > d4) prevlineasm1(vince[0],palookupoffse[0],y2ve[0]-d4-1,vplce[0],bufplce[0],i+0);
-        if (y2ve[1] > d4) prevlineasm1(vince[1],palookupoffse[1],y2ve[1]-d4-1,vplce[1],bufplce[1],i+1);
-        if (y2ve[2] > d4) prevlineasm1(vince[2],palookupoffse[2],y2ve[2]-d4-1,vplce[2],bufplce[2],i+2);
-        if (y2ve[3] > d4) prevlineasm1(vince[3],palookupoffse[3],y2ve[3]-d4-1,vplce[3],bufplce[3],i+3);
+        p = x+frameoffset+ylookup[d4+1];
+        if (y2ve[0] > d4) prevlineasm1(vince[0],palookupoffse[0],y2ve[0]-d4-1,vplce[0],bufplce[0],p+0);
+        if (y2ve[1] > d4) prevlineasm1(vince[1],palookupoffse[1],y2ve[1]-d4-1,vplce[1],bufplce[1],p+1);
+        if (y2ve[2] > d4) prevlineasm1(vince[2],palookupoffse[2],y2ve[2]-d4-1,vplce[2],bufplce[2],p+2);
+        if (y2ve[3] > d4) prevlineasm1(vince[3],palookupoffse[3],y2ve[3]-d4-1,vplce[3],bufplce[3],p+3);
     }
     for (; x<=x2; x++)
     {
@@ -7045,7 +7047,7 @@ static void dorotatesprite(int32_t sx, int32_t sy, int32_t z, int16_t a, int16_t
                     if (u4 > y1ve[2]) vplce[2] = prevlineasm1(vince[2],palookupoffse[2],u4-y1ve[2]-1,vplce[2],bufplce[2],ylookup[y1ve[2]]+p+2);
                     if (u4 > y1ve[3]) vplce[3] = prevlineasm1(vince[3],palookupoffse[3],u4-y1ve[3]-1,vplce[3],bufplce[3],ylookup[y1ve[3]]+p+3);
 
-                    if (d4 >= u4) vlineasm4(d4-u4+1,ylookup[u4]+p);
+                    if (d4 >= u4) vlineasm4(d4-u4+1, (char *)ylookup[u4]+p);
 
                     i = p+ylookup[d4+1];
                     if (y2ve[0] > d4) prevlineasm1(vince[0],palookupoffse[0],y2ve[0]-d4-1,vplce[0],bufplce[0],i+0);
@@ -7069,7 +7071,7 @@ static void dorotatesprite(int32_t sx, int32_t sy, int32_t z, int16_t a, int16_t
                     if (u4 > y1ve[2]) vplce[2] = mvlineasm1(vince[2],palookupoffse[2],u4-y1ve[2]-1,vplce[2],bufplce[2],ylookup[y1ve[2]]+p+2);
                     if (u4 > y1ve[3]) vplce[3] = mvlineasm1(vince[3],palookupoffse[3],u4-y1ve[3]-1,vplce[3],bufplce[3],ylookup[y1ve[3]]+p+3);
 
-                    if (d4 >= u4) mvlineasm4(d4-u4+1,ylookup[u4]+p);
+                    if (d4 >= u4) mvlineasm4(d4-u4+1, (char *)ylookup[u4]+p);
 
                     i = p+ylookup[d4+1];
                     if (y2ve[0] > d4) mvlineasm1(vince[0],palookupoffse[0],y2ve[0]-d4-1,vplce[0],bufplce[0],i+0);