diff --git a/src/main/fragment/vduz1=vduz2_plus_1.asm b/src/main/fragment/vduz1=vduz2_plus_1.asm
new file mode 100644
index 000000000..4925d2818
--- /dev/null
+++ b/src/main/fragment/vduz1=vduz2_plus_1.asm
@@ -0,0 +1,13 @@
+lda {z2}
+clc
+adc #1
+sta {z1}
+lda {z2}+1
+adc #0
+sta {z1}+1
+lda {z2}+2
+adc #0
+sta {z1}+2
+lda {z2}+3
+adc #0
+sta {z1}+3
diff --git a/src/main/kc/stdlib/fastmultiply.kc b/src/main/kc/stdlib/fastmultiply.kc
index 3adcb1743..36b2c297c 100644
--- a/src/main/kc/stdlib/fastmultiply.kc
+++ b/src/main/kc/stdlib/fastmultiply.kc
@@ -171,3 +171,17 @@ _dd:    lda #0
     }
     return *memR;
 }
+
+// Fast multiply two signed words to a signed double word result
+// Done in assembler to utilize fast addition A+X
+signed dword mulf16s(signed word a, signed word b) {
+    dword m = mulf16u((word)a, (word)b);
+    if(a<0) {
+        >m = (>m)-(word)b;
+    }
+    if(b<0) {
+        >m = (>m)-(word)a;
+        m = m + 1;
+    }
+    return (signed dword)m;
+}
diff --git a/src/test/kc/test-multiply-16bit.kc b/src/test/kc/test-multiply-16bit.kc
index eb63ff652..c7e05f6bb 100644
--- a/src/test/kc/test-multiply-16bit.kc
+++ b/src/test/kc/test-multiply-16bit.kc
@@ -97,13 +97,17 @@ void mul16s_compare() {
             b=b+4093;
             signed dword ms = muls16s(a, b);
             signed dword mn = mul16s(a,b);
+            signed dword mf = mulf16s(a,b);
             byte ok = 1;
+            if(ms!=mf) {
+                ok = 0;
+            }
             if(ms!=mn) {
                 ok = 0;
             }
             if(ok==0) {
                 *BGCOL = 2;
-                mul16s_error(a,b, ms, mn);
+                mul16s_error(a,b, ms, mn, mf);
                 return;
             }
         }
@@ -113,7 +117,7 @@ void mul16s_compare() {
     print_ln();
 }
 
-void mul16s_error(signed word a, signed word b, signed dword ms, signed dword mn) {
+void mul16s_error(signed word a, signed word b, signed dword ms, signed dword mn, signed dword mf) {
   print_str("signed word multiply mismatch @");
   print_sword(a);
   print_str("*@");
@@ -122,5 +126,7 @@ void mul16s_error(signed word a, signed word b, signed dword ms, signed dword mn
   print_sdword(ms);
   print_str(" / normal:@");
   print_sdword(mn);
+  print_str(" / fast:@");
+  print_sdword(mf);
   print_ln();
 }
diff --git a/src/test/ref/test-multiply-16bit.cfg b/src/test/ref/test-multiply-16bit.cfg
index 24cb04d96..f65c18afe 100644
--- a/src/test/ref/test-multiply-16bit.cfg
+++ b/src/test/ref/test-multiply-16bit.cfg
@@ -1,13 +1,13 @@
 @begin: scope:[]  from
   [0] phi() [ ] ( )
-  to:@30
-@30: scope:[]  from @begin
+  to:@31
+@31: scope:[]  from @begin
   [1] phi() [ ] ( )
   [2] call main  [ ] ( )
   to:@end
-@end: scope:[]  from @30
+@end: scope:[]  from @31
   [3] phi() [ ] ( )
-main: scope:[main]  from @30
+main: scope:[main]  from @31
   [4] *((const byte*) BGCOL#0) ← (byte/signed byte/word/signed word/dword/signed dword) 5 [ ] ( main:2 [ ] )
   [5] call print_cls  [ ] ( main:2 [ ] )
   to:main::@1
diff --git a/src/test/ref/test-multiply-16bit.log b/src/test/ref/test-multiply-16bit.log
index 56caf4e68..991fc87a8 100644
--- a/src/test/ref/test-multiply-16bit.log
+++ b/src/test/ref/test-multiply-16bit.log
@@ -348,7 +348,7 @@ mul16s::@return: scope:[mul16s]  from mul16s::@2
   (byte[512]) mulf_sqr1_hi#0 ← { fill( 512, 0) }
   (byte[512]) mulf_sqr2_lo#0 ← { fill( 512, 0) }
   (byte[512]) mulf_sqr2_hi#0 ← { fill( 512, 0) }
-  to:@23
+  to:@24
 mulf_init: scope:[mulf_init]  from main::@1
   (word) mulf_init::sqr#0 ← (byte/signed byte/word/signed word/dword/signed dword) 0
   (byte) mulf_init::x_2#0 ← (byte/signed byte/word/signed word/dword/signed dword) 0
@@ -462,17 +462,17 @@ mulf16u::@return: scope:[mulf16u]  from mulf16u
   (dword) mulf16u::return#1 ← (dword) mulf16u::return#3
   return 
   to:@return
-@23: scope:[]  from @19
+@24: scope:[]  from @19
   (byte*) print_screen#6 ← phi( @19/(byte*) print_screen#7 )
   (byte*) print_char_cursor#153 ← phi( @19/(byte*) print_char_cursor#154 )
   (byte*) print_line_cursor#61 ← phi( @19/(byte*) print_line_cursor#62 )
   (byte*) BGCOL#0 ← ((byte*)) (word/dword/signed dword) 53281
-  to:@30
-main: scope:[main]  from @30
-  (byte*) print_char_cursor#133 ← phi( @30/(byte*) print_char_cursor#143 )
-  (byte*) print_line_cursor#44 ← phi( @30/(byte*) print_line_cursor#54 )
-  (byte*) print_screen#4 ← phi( @30/(byte*) print_screen#5 )
-  (byte*) BGCOL#1 ← phi( @30/(byte*) BGCOL#4 )
+  to:@31
+main: scope:[main]  from @31
+  (byte*) print_char_cursor#133 ← phi( @31/(byte*) print_char_cursor#143 )
+  (byte*) print_line_cursor#44 ← phi( @31/(byte*) print_line_cursor#54 )
+  (byte*) print_screen#4 ← phi( @31/(byte*) print_screen#5 )
+  (byte*) BGCOL#1 ← phi( @31/(byte*) BGCOL#4 )
   *((byte*) BGCOL#1) ← (byte/signed byte/word/signed word/dword/signed dword) 5
   call print_cls 
   to:main::@1
@@ -1233,28 +1233,28 @@ mul16s_error::@return: scope:[mul16s_error]  from mul16s_error::@9
   (byte*) print_line_cursor#20 ← (byte*) print_line_cursor#41
   return 
   to:@return
-@30: scope:[]  from @23
-  (byte*) print_screen#5 ← phi( @23/(byte*) print_screen#6 )
-  (byte*) print_char_cursor#143 ← phi( @23/(byte*) print_char_cursor#153 )
-  (byte*) print_line_cursor#54 ← phi( @23/(byte*) print_line_cursor#61 )
-  (byte*) BGCOL#4 ← phi( @23/(byte*) BGCOL#0 )
+@31: scope:[]  from @24
+  (byte*) print_screen#5 ← phi( @24/(byte*) print_screen#6 )
+  (byte*) print_char_cursor#143 ← phi( @24/(byte*) print_char_cursor#153 )
+  (byte*) print_line_cursor#54 ← phi( @24/(byte*) print_line_cursor#61 )
+  (byte*) BGCOL#4 ← phi( @24/(byte*) BGCOL#0 )
   call main 
-  to:@31
-@31: scope:[]  from @30
-  (byte*) print_char_cursor#123 ← phi( @30/(byte*) print_char_cursor#27 )
-  (byte*) print_line_cursor#42 ← phi( @30/(byte*) print_line_cursor#8 )
+  to:@32
+@32: scope:[]  from @31
+  (byte*) print_char_cursor#123 ← phi( @31/(byte*) print_char_cursor#27 )
+  (byte*) print_line_cursor#42 ← phi( @31/(byte*) print_line_cursor#8 )
   (byte*) print_line_cursor#21 ← (byte*) print_line_cursor#42
   (byte*) print_char_cursor#62 ← (byte*) print_char_cursor#123
   to:@end
-@end: scope:[]  from @31
+@end: scope:[]  from @32
 
 SYMBOL TABLE SSA
 (const string) $0 = (string) "0123456789abcdef"
 (label) @10
 (label) @19
-(label) @23
-(label) @30
+(label) @24
 (label) @31
+(label) @32
 (label) @begin
 (label) @end
 (byte*) BGCOL
@@ -2816,7 +2816,7 @@ Culled Empty Block (label) print_cls::@2
 Culled Empty Block (label) mul16u::@3
 Culled Empty Block (label) @19
 Culled Empty Block (label) mulf_init::@6
-Culled Empty Block (label) @23
+Culled Empty Block (label) @24
 Culled Empty Block (label) main::@4
 Culled Empty Block (label) muls16u::@3
 Culled Empty Block (label) muls16s::@1
@@ -2830,7 +2830,7 @@ Culled Empty Block (label) mul16s_compare::@10
 Culled Empty Block (label) mul16s_compare::@13
 Culled Empty Block (label) mul16s_compare::@16
 Culled Empty Block (label) mul16s_error::@9
-Culled Empty Block (label) @31
+Culled Empty Block (label) @32
 Successful SSA optimization Pass2CullEmptyBlocks
 Self Phi Eliminated (byte*) BGCOL#21
 Self Phi Eliminated (byte*) print_line_cursor#89
@@ -2967,7 +2967,7 @@ Added new block during phi lifting mulf_init::@11(between mulf_init::@4 and mulf
 Added new block during phi lifting mulf_init::@12(between mulf_init::@3 and mulf_init::@4)
 Added new block during phi lifting print_cls::@3(between print_cls::@1 and print_cls::@1)
 Adding NOP phi() at start of @begin
-Adding NOP phi() at start of @30
+Adding NOP phi() at start of @31
 Adding NOP phi() at start of @end
 Adding NOP phi() at start of main::@1
 Adding NOP phi() at start of main::@2
@@ -3143,7 +3143,7 @@ Culled Empty Block (label) mulf_init::@9
 Culled Empty Block (label) mulf_init::@10
 Culled Empty Block (label) print_cls::@3
 Adding NOP phi() at start of @begin
-Adding NOP phi() at start of @30
+Adding NOP phi() at start of @31
 Adding NOP phi() at start of @end
 Adding NOP phi() at start of main::@1
 Adding NOP phi() at start of main::@2
@@ -3176,14 +3176,14 @@ Adding NOP phi() at start of print_cls
 FINAL CONTROL FLOW GRAPH
 @begin: scope:[]  from
   [0] phi() [ ] ( )
-  to:@30
-@30: scope:[]  from @begin
+  to:@31
+@31: scope:[]  from @begin
   [1] phi() [ ] ( )
   [2] call main  [ ] ( )
   to:@end
-@end: scope:[]  from @30
+@end: scope:[]  from @31
   [3] phi() [ ] ( )
-main: scope:[main]  from @30
+main: scope:[main]  from @31
   [4] *((const byte*) BGCOL#0) ← (byte/signed byte/word/signed word/dword/signed dword) 5 [ ] ( main:2 [ ] )
   [5] call print_cls  [ ] ( main:2 [ ] )
   to:main::@1
@@ -4256,15 +4256,15 @@ INITIAL ASM
   .label print_line_cursor = 9
 //SEG2 @begin
 bbegin:
-//SEG3 [1] phi from @begin to @30 [phi:@begin->@30]
-b30_from_bbegin:
-  jmp b30
-//SEG4 @30
-b30:
+//SEG3 [1] phi from @begin to @31 [phi:@begin->@31]
+b31_from_bbegin:
+  jmp b31
+//SEG4 @31
+b31:
 //SEG5 [2] call main  [ ] ( )
   jsr main
-//SEG6 [3] phi from @30 to @end [phi:@30->@end]
-bend_from_b30:
+//SEG6 [3] phi from @31 to @end [phi:@31->@end]
+bend_from_b31:
   jmp bend
 //SEG7 @end
 bend:
@@ -6920,15 +6920,15 @@ ASSEMBLER BEFORE OPTIMIZATION
   .label print_line_cursor = 6
 //SEG2 @begin
 bbegin:
-//SEG3 [1] phi from @begin to @30 [phi:@begin->@30]
-b30_from_bbegin:
-  jmp b30
-//SEG4 @30
-b30:
+//SEG3 [1] phi from @begin to @31 [phi:@begin->@31]
+b31_from_bbegin:
+  jmp b31
+//SEG4 @31
+b31:
 //SEG5 [2] call main  [ ] ( )
   jsr main
-//SEG6 [3] phi from @30 to @end [phi:@30->@end]
-bend_from_b30:
+//SEG6 [3] phi from @31 to @end [phi:@31->@end]
+bend_from_b31:
   jmp bend
 //SEG7 @end
 bend:
@@ -8882,7 +8882,7 @@ print_cls: {
   mulf_sqr2_hi: .fill $200, 0
 
 ASSEMBLER OPTIMIZATIONS
-Removing instruction jmp b30
+Removing instruction jmp b31
 Removing instruction jmp bend
 Removing instruction jmp b1
 Removing instruction jmp b2
@@ -9034,8 +9034,8 @@ Replacing label b3_from_b4 with b3
 Replacing label b1_from_b1 with b1
 Replacing label b1_from_b1 with b1
 Removing instruction bbegin:
-Removing instruction b30_from_bbegin:
-Removing instruction bend_from_b30:
+Removing instruction b31_from_bbegin:
+Removing instruction bend_from_b31:
 Removing instruction b1_from_main:
 Removing instruction mulf_init_from_b1:
 Removing instruction b2_from_b1:
@@ -9113,7 +9113,7 @@ Removing instruction b12_from_b3:
 Removing instruction b4_from_b12:
 Removing instruction b1_from_b1:
 Succesful ASM optimization Pass5RedundantLabelElimination
-Removing instruction b30:
+Removing instruction b31:
 Removing instruction bend:
 Removing instruction print_cls_from_main:
 Removing instruction b1:
@@ -9257,7 +9257,7 @@ Succesful ASM optimization Pass5UnreachableCodeElimination
 Fixing long branch [596] bne b1 to beq
 
 FINAL SYMBOL TABLE
-(label) @30
+(label) @31
 (label) @begin
 (label) @end
 (byte*) BGCOL
@@ -9689,11 +9689,11 @@ Score: 441007
   .label print_char_cursor = $e
   .label print_line_cursor = 6
 //SEG2 @begin
-//SEG3 [1] phi from @begin to @30 [phi:@begin->@30]
-//SEG4 @30
+//SEG3 [1] phi from @begin to @31 [phi:@begin->@31]
+//SEG4 @31
 //SEG5 [2] call main  [ ] ( )
   jsr main
-//SEG6 [3] phi from @30 to @end [phi:@30->@end]
+//SEG6 [3] phi from @31 to @end [phi:@31->@end]
 //SEG7 @end
 //SEG8 main
 main: {
diff --git a/src/test/ref/test-multiply-16bit.sym b/src/test/ref/test-multiply-16bit.sym
index 206b30956..c32bd319a 100644
--- a/src/test/ref/test-multiply-16bit.sym
+++ b/src/test/ref/test-multiply-16bit.sym
@@ -1,4 +1,4 @@
-(label) @30
+(label) @31
 (label) @begin
 (label) @end
 (byte*) BGCOL