diff --git a/compiler/examples/comparison_ifs_byte.p8 b/compiler/examples/comparison_ifs_byte.p8
index 9d8e9e386..8c495a4df 100644
--- a/compiler/examples/comparison_ifs_byte.p8
+++ b/compiler/examples/comparison_ifs_byte.p8
@@ -10,7 +10,7 @@
         byte v1
         byte v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         v1 = 100
@@ -110,7 +110,7 @@
         else
             c64.STROUT("error in 22>=22!\n")
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
     }
diff --git a/compiler/examples/comparison_ifs_float.p8 b/compiler/examples/comparison_ifs_float.p8
index c338d9dd8..49d7acde2 100644
--- a/compiler/examples/comparison_ifs_float.p8
+++ b/compiler/examples/comparison_ifs_float.p8
@@ -10,7 +10,7 @@
         float v1
         float v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         v1 = 1.11
@@ -110,7 +110,7 @@
         else
             c64.STROUT("error in -22.2>=-22.2!\n")
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
     }
diff --git a/compiler/examples/comparison_ifs_uword.p8 b/compiler/examples/comparison_ifs_uword.p8
index d5a21e51c..9dd7fe07b 100644
--- a/compiler/examples/comparison_ifs_uword.p8
+++ b/compiler/examples/comparison_ifs_uword.p8
@@ -10,7 +10,7 @@
         uword v1
         uword v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         v1 = 100
@@ -110,7 +110,7 @@
         else
             c64.STROUT("error in 322>=322!\n")
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
     }
diff --git a/compiler/examples/comparisons_byte.p8 b/compiler/examples/comparisons_byte.p8
index 2676f7335..2a43c6bd7 100644
--- a/compiler/examples/comparisons_byte.p8
+++ b/compiler/examples/comparisons_byte.p8
@@ -11,7 +11,7 @@
 
         ; check stack usage:
         c64.STROUT("signed byte ")
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT(' ')
 
         cr=v1==v2
@@ -39,7 +39,7 @@
         cr=v1>=v2
         cr=v1>=v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         ; comparisons:
@@ -58,7 +58,7 @@
         c64.STROUT("v1=20, v2=-111\n")
         compare()
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
         return
 
diff --git a/compiler/examples/comparisons_float.p8 b/compiler/examples/comparisons_float.p8
index dd8a3e2a6..fd3c295e9 100644
--- a/compiler/examples/comparisons_float.p8
+++ b/compiler/examples/comparisons_float.p8
@@ -13,7 +13,7 @@
 
         ; check stack usage:
         c64.STROUT("floating point ")
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT(' ')
 
         cr=v1==v2
@@ -41,7 +41,7 @@
         cr=v1>=v2
         cr=v1>=v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         ; comparisons:
@@ -75,7 +75,7 @@
         c64.STROUT("v1 = v2 = 0\n")
         compare()
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         return
diff --git a/compiler/examples/comparisons_ubyte.p8 b/compiler/examples/comparisons_ubyte.p8
index a6249e577..e6451dbe9 100644
--- a/compiler/examples/comparisons_ubyte.p8
+++ b/compiler/examples/comparisons_ubyte.p8
@@ -11,7 +11,7 @@
 
         ; check stack usage:
         c64.STROUT("unsigned byte ")
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT(' ')
 
         cr=v1==v2
@@ -39,7 +39,7 @@
         cr=v1>=v2
         cr=v1>=v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         ; comparisons:
@@ -58,7 +58,7 @@
         c64.STROUT("v1=220, v2=10\n")
         compare()
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         return
diff --git a/compiler/examples/comparisons_uword.p8 b/compiler/examples/comparisons_uword.p8
index d054ec465..a886e2a65 100644
--- a/compiler/examples/comparisons_uword.p8
+++ b/compiler/examples/comparisons_uword.p8
@@ -11,7 +11,7 @@
 
         ; check stack usage:
         c64.STROUT("unsigned word ")
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT(' ')
 
         cr=v1==v2
@@ -39,7 +39,7 @@
         cr=v1>=v2
         cr=v1>=v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         ; comparisons:
@@ -88,7 +88,7 @@
         c64.STROUT("v1 = v2 = aa\n")
         compare()
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         return
diff --git a/compiler/examples/comparisons_word.p8 b/compiler/examples/comparisons_word.p8
index 011f8685f..21723dae9 100644
--- a/compiler/examples/comparisons_word.p8
+++ b/compiler/examples/comparisons_word.p8
@@ -12,7 +12,7 @@
 
         ; check stack usage:
         c64.STROUT("signed word ")
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT(' ')
 
         cr=v1==v2
@@ -40,7 +40,7 @@
         cr=v1>=v2
         cr=v1>=v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         ; comparisons:
@@ -125,7 +125,7 @@
         c64.STROUT("v1 = v2 = aa\n")
         compare()
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         return
diff --git a/compiler/examples/comparsion_ifs_ubyte.p8 b/compiler/examples/comparsion_ifs_ubyte.p8
index 3d440ec2a..2af1a22a9 100644
--- a/compiler/examples/comparsion_ifs_ubyte.p8
+++ b/compiler/examples/comparsion_ifs_ubyte.p8
@@ -10,7 +10,7 @@
         ubyte v1
         ubyte v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         v1 = 100
@@ -110,7 +110,7 @@
         else
             c64.STROUT("error in 22>=22!\n")
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
     }
diff --git a/compiler/examples/comparsion_ifs_word.p8 b/compiler/examples/comparsion_ifs_word.p8
index 747eca9e8..ca4d1a717 100644
--- a/compiler/examples/comparsion_ifs_word.p8
+++ b/compiler/examples/comparsion_ifs_word.p8
@@ -10,7 +10,7 @@
         word v1
         word v2
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
         v1 = 100
@@ -110,7 +110,7 @@
         else
             c64.STROUT("error in -222>=-222!\n")
 
-        c64scr.print_byte_decimal(X)
+        c64scr.print_byte(X)
         c64.CHROUT('\n')
 
     }
diff --git a/compiler/examples/numbergame-c64.p8 b/compiler/examples/numbergame-c64.p8
index 4b1f40181..053c89a37 100644
--- a/compiler/examples/numbergame-c64.p8
+++ b/compiler/examples/numbergame-c64.p8
@@ -39,7 +39,7 @@
 
 ask_guess:
         c64.STROUT("\nYou have ")
-        c64scr.print_ubyte_decimal(attempts_left)
+        c64scr.print_ubyte(attempts_left)
         c64.STROUT(" guess")
         if(attempts_left>1)
             c64.STROUT("es")
@@ -66,7 +66,7 @@ ask_guess:
 
         ; game over.
         c64.STROUT("\nToo bad! It was: ")
-        c64scr.print_ubyte_decimal(secretnumber)
+        c64scr.print_ubyte(secretnumber)
         c64.CHROUT('\n')
 
 goodbye:
diff --git a/compiler/examples/numbergame-novm.p8 b/compiler/examples/numbergame-novm.p8
index aeccbd34f..a8cda640c 100644
--- a/compiler/examples/numbergame-novm.p8
+++ b/compiler/examples/numbergame-novm.p8
@@ -22,11 +22,11 @@
         for ubyte attempts_left in 10 to 1 step -1 {
 
 ; stackptr debugging
-;            c64scr.print_byte_decimal(X)
+;            c64scr.print_byte(X)
 ;            c64.CHROUT('\n')
 
             c64scr.print_string("\nYou have ")
-            c64scr.print_ubyte_decimal(attempts_left)
+            c64scr.print_ubyte(attempts_left)
             c64scr.print_string(" guess")
             if attempts_left>1
                 c64scr.print_string("es")
@@ -36,15 +36,15 @@
 
 ; debug info
 ;            c64scr.print_string(" > attempts left=")
-;            c64scr.print_byte_decimal(attempts_left)
+;            c64scr.print_byte(attempts_left)
 ;            c64scr.print_string("\n > secretnumber=")
-;            c64scr.print_byte_decimal(secretnumber)
+;            c64scr.print_byte(secretnumber)
 ;            c64scr.print_string("\n > input=")
 ;            c64scr.print_string(input)
 ;            c64scr.print_string("\n > guess=")
-;            c64scr.print_byte_decimal(guess)
+;            c64scr.print_byte(guess)
 ;            c64.CHROUT('\n')
-;            c64scr.print_byte_decimal(X)    ; stackptr debugging
+;            c64scr.print_byte(X)    ; stackptr debugging
 ;            c64.CHROUT('\n')
 
             if guess==secretnumber {
@@ -66,7 +66,7 @@
                 c64scr.print_string("\n\nYou guessed it, impressive!\n")
             else {
                 c64scr.print_string("\nToo bad! My number was: ")
-                c64scr.print_ubyte_decimal(secretnumber)
+                c64scr.print_ubyte(secretnumber)
                 c64scr.print_string(".\n")
             }
             c64scr.print_string("Thanks for playing, ")
diff --git a/compiler/examples/test.p8 b/compiler/examples/test.p8
index d39094578..94b94ff51 100644
--- a/compiler/examples/test.p8
+++ b/compiler/examples/test.p8
@@ -3,24 +3,69 @@
 
 ~ main {
 
-    memory ubyte[40] screen    = $0400
-
-    poke()
     sub start()  {
-        float t = 0.0
-        while(true) {
-            ubyte x = lsb(round(sin(t)*15.0))+20
-            ubyte y = lsb(round(cos(t)*10.0))+12
 
-;            vm_gfx_text(x, y, 1, "*")
-            ;vm_gfx_pixel(x,y,1)
-;            c64scr.print_ubyte_decimal(x)
-;            c64.CHROUT(',')
-;            c64scr.print_ubyte_decimal(y)
-;            c64.CHROUT('\n')
-            screen[x] = '*'
-            t+=0.1
-        }
+        uword   uw1
+        uword   uw2
+        word    w1
+        word    w2
+        ubyte   ub1
+        ubyte   ub2
+        byte    b1
+        byte    b2
+        float   f1
+        float   f2
+
+        ub1 = 200
+        c64scr.print_ubyte(abs(ub1))    ; ok
+        c64.CHROUT('\n')
+        b1 = 100
+        c64scr.print_byte(abs(b1))  ; ok
+        c64.CHROUT('\n')
+        b1 = 0
+        c64scr.print_byte(abs(b1))  ;ok
+        c64.CHROUT('\n')
+        b1 = -100
+        c64scr.print_byte(abs(b1))  ;ok
+        c64.CHROUT('\n')
+
+        uw1 = 1000
+        c64scr.print_uword(abs(uw1))    ;ok
+        c64.CHROUT('\n')
+        w1 = 0
+        c64scr.print_word(abs(w1))  ;ok
+        c64.CHROUT('\n')
+        w1 = -1000
+        c64scr.print_word(abs(w1))  ;ok
+        c64.CHROUT('\n')
+
+        f1 = 11.22
+        c64flt.print_float(abs(f1)) ;ok
+        c64.CHROUT('\n')
+        f1 = 0
+        c64flt.print_float(abs(f1)) ;ok
+        c64.CHROUT('\n')
+        f1 = -22.33
+        c64flt.print_float(abs(f1)) ;@todo FAIL
+        c64.CHROUT('\n')
+
+;        float t = 0.0
+;        ubyte color=0
+;        while(true) {
+;            ubyte x = lsb(round(sin(t)*17.0))+20
+;            ubyte y = lsb(round(cos(t*1.3634567)*10.0))+12
+;
+;            c64scr.setchrclr(x,y,color,color)
+;            color++
+;;            vm_gfx_text(x, y, 1, "*")
+;            ;vm_gfx_pixel(x,y,1)
+;;            c64scr.print_ubyte(x)
+;;            c64.CHROUT(',')
+;;            c64scr.print_ubyte(y)
+;;            c64.CHROUT('\n')
+;            ;screen[x] = '*'
+;            t+=0.1
+;        }
     }
 }
 
diff --git a/compiler/src/prog8/compiler/Compiler.kt b/compiler/src/prog8/compiler/Compiler.kt
index cc4de66f5..79a76ca06 100644
--- a/compiler/src/prog8/compiler/Compiler.kt
+++ b/compiler/src/prog8/compiler/Compiler.kt
@@ -711,6 +711,17 @@ private class StatementTranslator(private val prog: IntermediateProgram,
                 }
 
             }
+            "abs" -> {
+                // 1 argument, type determines the exact opcode to use
+                val arg = args.single()
+                when (arg.resultingDatatype(namespace, heap)) {
+                    DataType.UBYTE, DataType.UWORD -> {}
+                    DataType.BYTE -> prog.instr(Opcode.ABS_B)
+                    DataType.WORD -> prog.instr(Opcode.ABS_W)
+                    DataType.FLOAT -> prog.instr(Opcode.ABS_F)
+                    else -> throw CompilerException("wrong datatype for $funcname()")
+                }
+            }
             "flt" -> {
                 // 1 argument, type determines the exact opcode to use
                 val arg = args.single()
diff --git a/compiler/src/prog8/compiler/intermediate/Opcode.kt b/compiler/src/prog8/compiler/intermediate/Opcode.kt
index 882e49b6c..7d329053a 100644
--- a/compiler/src/prog8/compiler/intermediate/Opcode.kt
+++ b/compiler/src/prog8/compiler/intermediate/Opcode.kt
@@ -72,6 +72,9 @@ enum class Opcode {
     NEG_B,
     NEG_W,
     NEG_F,
+    ABS_B,
+    ABS_W,
+    ABS_F,
 
     // bit shifts and bitwise arithmetic
     SHL_BYTE,
diff --git a/compiler/src/prog8/compiler/intermediate/Value.kt b/compiler/src/prog8/compiler/intermediate/Value.kt
index 38002a3be..971d2b3f7 100644
--- a/compiler/src/prog8/compiler/intermediate/Value.kt
+++ b/compiler/src/prog8/compiler/intermediate/Value.kt
@@ -325,6 +325,15 @@ class Value(val type: DataType, numericvalueOrHeapId: Number) {
         }
     }
 
+    fun abs(): Value {
+        return when(type) {
+            DataType.BYTE -> Value(DataType.BYTE, abs(byteval!!.toInt()))
+            DataType.WORD -> Value(DataType.WORD, abs(wordval!!))
+            DataType.FLOAT -> Value(DataType.FLOAT, abs(floatval!!))
+            else -> throw VmExecutionException("abs can only work on byte/word/float")
+        }
+    }
+
     fun bitand(other: Value): Value {
         val v1 = integerValue()
         val v2 = other.integerValue()
diff --git a/compiler/src/prog8/compiler/target/c64/AsmGen.kt b/compiler/src/prog8/compiler/target/c64/AsmGen.kt
index 97ff64f83..0c0b65812 100644
--- a/compiler/src/prog8/compiler/target/c64/AsmGen.kt
+++ b/compiler/src/prog8/compiler/target/c64/AsmGen.kt
@@ -409,16 +409,7 @@ class AsmGen(val options: CompilationOptions, val program: IntermediateProgram,
             Opcode.RSAVE -> {
                 // save cpu status flag and all registers A, X, Y.
                 // see http://6502.org/tutorials/register_preservation.html
-                """
-                php
-                sta  ${C64Zeropage.SCRATCH_REG}
-                pha
-                txa
-                pha
-                tya
-                pha
-                lda  ${C64Zeropage.SCRATCH_REG}
-                """
+                " php |  sta  ${C64Zeropage.SCRATCH_REG} | pha  | txa  | pha  | tya  | pha  | lda  ${C64Zeropage.SCRATCH_REG}"
             }
             Opcode.RRESTORE -> {
                 // restore all registers and cpu status flag
@@ -623,16 +614,12 @@ class AsmGen(val options: CompilationOptions, val program: IntermediateProgram,
                 jsr  prog8_lib.dec_var_f
                 """
             }
-            Opcode.NEG_B -> {
-                """
-                lda  ${(ESTACK_LO+1).toHex()},x
-                eor  #255
-                sec
-                adc  #0
-                sta  ${(ESTACK_LO+1).toHex()},x
-                """
-            }
+            Opcode.NEG_B -> " jsr  prog8_lib.neg_b"
+            Opcode.NEG_W -> " jsr  prog8_lib.neg_w"
             Opcode.NEG_F -> " jsr  prog8_lib.neg_f"
+            Opcode.ABS_B -> " jsr  prog8_lib.abs_b"
+            Opcode.ABS_W -> " jsr  prog8_lib.abs_w"
+            Opcode.ABS_F -> " jsr  prog8_lib.abs_f"
             Opcode.INV_BYTE -> {
                 """
                 lda  ${(ESTACK_LO + 1).toHex()},x
@@ -640,38 +627,9 @@ class AsmGen(val options: CompilationOptions, val program: IntermediateProgram,
                 sta  ${(ESTACK_LO + 1).toHex()},x
                 """
             }
-            Opcode.INV_WORD -> {
-                """
-                lda  ${(ESTACK_LO+1).toHex()},x
-                eor  #255
-                sta  ${(ESTACK_LO+1).toHex()},x
-                lda  ${(ESTACK_HI+1).toHex()},x
-                eor  #255
-                sta  ${(ESTACK_HI+1).toHex()},x
-                """
-            }
-            Opcode.NOT_BYTE -> {
-                """
-                lda  ${(ESTACK_LO+1).toHex()},x
-                beq  +
-                lda  #0
-                beq ++
-+               lda  #1
-+               sta  ${(ESTACK_LO+1).toHex()},x
-                """
-            }
-            Opcode.NOT_WORD -> {
-                """
-                lda  ${(ESTACK_LO + 1).toHex()},x
-                ora  ${(ESTACK_HI + 1).toHex()},x
-                beq  +
-                lda  #0
-                beq  ++
-+               lda  #1
-+               sta  ${(ESTACK_LO + 1).toHex()},x |  sta  ${(ESTACK_HI + 1).toHex()},x
-                """
-            }
-
+            Opcode.INV_WORD -> " jsr  prog8_lib.inv_word"
+            Opcode.NOT_BYTE -> " jsr  prog8_lib.not_byte"
+            Opcode.NOT_WORD -> " jsr  prog8_lib.not_word"
             Opcode.BCS -> " bcs  ${ins.callLabel}"
             Opcode.BCC -> " bcc  ${ins.callLabel}"
             Opcode.BNEG -> " bmi  ${ins.callLabel}"
@@ -744,11 +702,9 @@ class AsmGen(val options: CompilationOptions, val program: IntermediateProgram,
                 sta  ${(ESTACK_LO + 1).toHex()},x
                 """
             }
-            Opcode.ADD_W -> "  jsr  prog8_lib.add_w"    // todo inline?
-            Opcode.ADD_UW -> "  jsr  prog8_lib.add_uw"  // todo inline?
+            Opcode.ADD_W, Opcode.ADD_UW -> "  jsr  prog8_lib.add_w"
             Opcode.ADD_F -> "  jsr  prog8_lib.add_f"
-            Opcode.SUB_W -> "  jsr  prog8_lib.sub_w"    // todo inline?
-            Opcode.SUB_UW -> "  jsr  prog8_lib.sub_uw"    // todo inline?
+            Opcode.SUB_W, Opcode.SUB_UW -> "  jsr  prog8_lib.sub_w"
             Opcode.SUB_F -> "  jsr  prog8_lib.sub_f"
             Opcode.MUL_B -> "  jsr  prog8_lib.mul_b"
             Opcode.MUL_UB -> "  jsr  prog8_lib.mul_ub"
diff --git a/compiler/src/prog8/functions/BuiltinFunctions.kt b/compiler/src/prog8/functions/BuiltinFunctions.kt
index 397b90df6..0f00ac894 100644
--- a/compiler/src/prog8/functions/BuiltinFunctions.kt
+++ b/compiler/src/prog8/functions/BuiltinFunctions.kt
@@ -22,10 +22,11 @@ val BuiltinFunctions = mapOf(
     "ror2"        to FunctionSignature(false, listOf(BuiltinFunctionParam("item", setOf(DataType.UBYTE, DataType.UWORD))), null),
     "lsl"         to FunctionSignature(false, listOf(BuiltinFunctionParam("item", setOf(DataType.UBYTE, DataType.UWORD))), null),
     "lsr"         to FunctionSignature(false, listOf(BuiltinFunctionParam("item", setOf(DataType.UBYTE, DataType.UWORD))), null),
-        // these few have a return value depending on the argument list:
+        // these few have a return value depending on the argument(s):
     "max"         to FunctionSignature(true, listOf(BuiltinFunctionParam("values", ArrayDatatypes)), null) { a, p, n, h -> collectionArgOutputNumber(a, p, n, h) { it.max()!! }},        // type depends on args
     "min"         to FunctionSignature(true, listOf(BuiltinFunctionParam("values", ArrayDatatypes)), null) { a, p, n, h -> collectionArgOutputNumber(a, p, n, h) { it.min()!! }},        // type depends on args
     "sum"         to FunctionSignature(true, listOf(BuiltinFunctionParam("values", ArrayDatatypes)), null) { a, p, n, h -> collectionArgOutputNumber(a, p, n, h) { it.sum() }},        // type depends on args
+    "abs"         to FunctionSignature(true, listOf(BuiltinFunctionParam("value", NumericDatatypes)), null, ::builtinAbs),      // type depends on argument
         // normal functions follow:
     "sin"         to FunctionSignature(true, listOf(BuiltinFunctionParam("rads", setOf(DataType.FLOAT))), DataType.FLOAT) { a, p, n, h -> oneDoubleArg(a, p, n, h, Math::sin) },
     "cos"         to FunctionSignature(true, listOf(BuiltinFunctionParam("rads", setOf(DataType.FLOAT))), DataType.FLOAT) { a, p, n, h -> oneDoubleArg(a, p, n, h, Math::cos) },
@@ -38,7 +39,6 @@ val BuiltinFunctions = mapOf(
     "rad"         to FunctionSignature(true, listOf(BuiltinFunctionParam("value", setOf(DataType.FLOAT))), DataType.FLOAT) { a, p, n, h -> oneDoubleArg(a, p, n, h, Math::toRadians) },
     "deg"         to FunctionSignature(true, listOf(BuiltinFunctionParam("value", setOf(DataType.FLOAT))), DataType.FLOAT) { a, p, n, h -> oneDoubleArg(a, p, n, h, Math::toDegrees) },
     "avg"         to FunctionSignature(true, listOf(BuiltinFunctionParam("values", ArrayDatatypes)), DataType.FLOAT, ::builtinAvg),
-    "abs"         to FunctionSignature(true, listOf(BuiltinFunctionParam("value", setOf(DataType.FLOAT))), DataType.FLOAT, ::builtinAbs),
     "round"       to FunctionSignature(true, listOf(BuiltinFunctionParam("value", setOf(DataType.FLOAT))), DataType.WORD) { a, p, n, h -> oneDoubleArgOutputWord(a, p, n, h, Math::round) },
     "floor"       to FunctionSignature(true, listOf(BuiltinFunctionParam("value", setOf(DataType.FLOAT))), DataType.WORD) { a, p, n, h -> oneDoubleArgOutputWord(a, p, n, h, Math::floor) },
     "ceil"        to FunctionSignature(true, listOf(BuiltinFunctionParam("value", setOf(DataType.FLOAT))), DataType.WORD) { a, p, n, h -> oneDoubleArgOutputWord(a, p, n, h, Math::ceil) },
@@ -128,7 +128,7 @@ fun builtinFunctionReturnType(function: String, args: List<IExpression>, namespa
     // function has return values, but the return type depends on the arguments
 
     return when (function) {
-        "max", "min" -> {
+        "max", "min", "abs" -> {
             val dt = datatypeFromListArg(args.single())
             when(dt) {
                 DataType.UBYTE, DataType.BYTE, DataType.UWORD, DataType.WORD, DataType.FLOAT -> dt
diff --git a/compiler/src/prog8/stackvm/StackVm.kt b/compiler/src/prog8/stackvm/StackVm.kt
index 4f4f766de..9109cb585 100644
--- a/compiler/src/prog8/stackvm/StackVm.kt
+++ b/compiler/src/prog8/stackvm/StackVm.kt
@@ -542,6 +542,21 @@ class StackVm(private var traceOutputFile: String?) {
                 checkDt(v, DataType.FLOAT)
                 evalstack.push(v.neg())
             }
+            Opcode.ABS_B -> {
+                val v = evalstack.pop()
+                checkDt(v, DataType.BYTE)
+                evalstack.push(v.abs())
+            }
+            Opcode.ABS_W -> {
+                val v = evalstack.pop()
+                checkDt(v, DataType.WORD)
+                evalstack.push(v.abs())
+            }
+            Opcode.ABS_F -> {
+                val v = evalstack.pop()
+                checkDt(v, DataType.FLOAT)
+                evalstack.push(v.abs())
+            }
             Opcode.SHL_BYTE -> {
                 val v = evalstack.pop()
                 checkDt(v, DataType.UBYTE)
diff --git a/prog8lib/c64utils.p8 b/prog8lib/c64utils.p8
index d11ca8cbc..1957c91c3 100644
--- a/prog8lib/c64utils.p8
+++ b/prog8lib/c64utils.p8
@@ -11,6 +11,10 @@
 
 ~ c64utils {
 
+		const   uword  ESTACK_LO	= $ce00
+		const   uword  ESTACK_HI	= $cf00
+		
+		
 ; ----- utility functions ----
 
 asmsub  init_system  () -> clobbers(A,X,Y) -> ()  {
@@ -613,7 +617,7 @@ asmsub  print_pstring  (text: str_p @ AY) -> clobbers(A,X) -> (ubyte @ Y)  {
 }
 
 
-asmsub  print_ubyte_decimal0  (value: ubyte @ A) -> clobbers(A,X,Y) -> ()  {
+asmsub  print_ubyte0  (value: ubyte @ A) -> clobbers(A,X,Y) -> ()  {
 	; ---- print the ubyte in A in decimal form, with left padding 0s (3 positions total)
 	%asm {{
 		jsr  c64utils.ubyte2decimal
@@ -628,7 +632,7 @@ asmsub  print_ubyte_decimal0  (value: ubyte @ A) -> clobbers(A,X,Y) -> ()  {
 }
 
 
-asmsub  print_ubyte_decimal  (value: ubyte @ A) -> clobbers(A,X,Y) -> ()  {
+asmsub  print_ubyte  (value: ubyte @ A) -> clobbers(A,X,Y) -> ()  {
 	; ---- print the ubyte in A in decimal form, without left padding 0s
 	%asm {{
 		jsr  c64utils.ubyte2decimal
@@ -649,7 +653,7 @@ _print_tens	txa
 	}}
 }
 	
-asmsub  print_byte_decimal  (value: byte @ A) -> clobbers(A,X,Y) -> ()  {
+asmsub  print_byte  (value: byte @ A) -> clobbers(A,X,Y) -> ()  {
 	; ---- print the byte in A in decimal form, without left padding 0s
 	%asm {{
 		pha
@@ -659,7 +663,7 @@ asmsub  print_byte_decimal  (value: byte @ A) -> clobbers(A,X,Y) -> ()  {
 		jsr  c64.CHROUT
 +		pla
 		jsr  c64utils.byte2decimal
-		jmp  print_ubyte_decimal._print_byte_digits
+		jmp  print_ubyte._print_byte_digits
 	}}
 }
 
@@ -694,7 +698,7 @@ asmsub print_uword_hex  (prefix: ubyte @ Pc, value: uword @ AY) -> clobbers(A,X,
 }
 
 
-asmsub  print_uword_decimal0  (value: uword @ AY) -> clobbers(A,X,Y) -> ()  {
+asmsub  print_uword0  (value: uword @ AY) -> clobbers(A,X,Y) -> ()  {
 	; ---- print the uword in A/Y in decimal form, with left padding 0s (5 positions total)
 	; @todo shorter in loop form?
 	%asm {{
@@ -713,7 +717,7 @@ asmsub  print_uword_decimal0  (value: uword @ AY) -> clobbers(A,X,Y) -> ()  {
 }
 
 
-asmsub  print_uword_decimal  (value: uword @ AY) -> clobbers(A,X,Y) -> ()  {
+asmsub  print_uword  (value: uword @ AY) -> clobbers(A,X,Y) -> ()  {
 	; ---- print the uword in A/Y in decimal form, without left padding 0s
 	%asm {{
 		jsr  c64utils.uword2decimal
@@ -745,7 +749,7 @@ _pr_decimal
 	}}
 }
 
-asmsub  print_word_decimal  (value: word @ AY) -> clobbers(A,X,Y) -> ()  {
+asmsub  print_word  (value: word @ AY) -> clobbers(A,X,Y) -> ()  {
 	; ---- print the (signed) word in A/Y in decimal form, without left padding 0s
 	%asm {{
 		cpy  #0
@@ -762,7 +766,7 @@ asmsub  print_word_decimal  (value: word @ AY) -> clobbers(A,X,Y) -> ()  {
 		adc  #1
 		bcc  +
 		iny
-+		jmp  print_uword_decimal
++		jmp  print_uword
 	}}
 }
 
@@ -787,4 +791,80 @@ asmsub  input_chars  (buffer: uword @ AY) -> clobbers(A, X) -> (ubyte @ Y)  {
 	}}
 }
 
+asmsub  setchr  (col: ubyte @Y, row: ubyte @A) -> clobbers(A) -> ()  {
+	; ---- set the character in SCRATCH_ZPB1 on the screen matrix at the given position
+	%asm {{
+		sty  c64.SCRATCH_ZPREG
+		asl  a
+		tay
+		lda  _screenrows+1,y
+		sta  _mod+2
+		lda  _screenrows,y
+		clc
+		adc  c64.SCRATCH_ZPREG
+		sta  _mod+1
+		bcc  +
+		inc  _mod+2
++		lda  c64.SCRATCH_ZPB1
+_mod		sta  $ffff		; modified
+		rts
+		
+_screenrows	.word  $0400 + range(0, 1000, 40)
+	}}
+}
+
+asmsub  setclr  (col: ubyte @Y, row: ubyte @A) -> clobbers(A) -> ()  {
+	; ---- set the color in SCRATCH_ZPB1 on the screen matrix at the given position
+	%asm {{
+		sty  c64.SCRATCH_ZPREG
+		asl  a
+		tay
+		lda  _colorrows+1,y
+		sta  _mod+2
+		lda  _colorrows,y
+		clc
+		adc  c64.SCRATCH_ZPREG
+		sta  _mod+1
+		bcc  +
+		inc  _mod+2
++		lda  c64.SCRATCH_ZPB1
+_mod		sta  $ffff		; modified
+		rts
+		
+_colorrows	.word  $d800 + range(0, 1000, 40)
+	}}
+}
+
+		
+sub  setchrclr  (column: ubyte, row: ubyte, char: ubyte, color: ubyte)  {
+	; ---- set char+color at the given position on the screen
+	%asm {{
+		lda  setchrclr_row
+		asl  a
+		tay
+		lda  setchr._screenrows+1,y
+		sta  _charmod+2
+		adc  #$d4
+		sta  _colormod+2
+		lda  setchr._screenrows,y
+		clc
+		adc  setchrclr_column
+		sta  _charmod+1
+		sta  _colormod+1
+		bcc  +
+		inc  _charmod+2
+		inc  _colormod+2
++		lda  setchrclr_char
+_charmod	sta  $ffff		; modified
+		lda  setchrclr_color
+_colormod	sta  $ffff		; modified
+		inx
+		inx
+		inx
+		inx
+		rts
+	}}	
+}
+
+
 }  ; ---- end block c64scr
diff --git a/prog8lib/prog8lib.p8 b/prog8lib/prog8lib.p8
index a2f60ea1e..86c7f772a 100644
--- a/prog8lib/prog8lib.p8
+++ b/prog8lib/prog8lib.p8
@@ -357,6 +357,26 @@ mul_f		.proc
 		jmp  push_fac1_as_result
 		.pend
 
+neg_b		.proc
+		lda  ESTACK_LO+1,x
+		eor  #255
+		clc
+		adc  #1
+		sta  ESTACK_LO+1,x
+		rts
+		.pend
+                
+neg_w		.proc
+		sec
+		lda  #0
+		sbc  ESTACK_LO+1,x
+		sta  ESTACK_LO+1,x
+		lda  #0
+		sbc  ESTACK_HI+1,x
+		sta  ESTACK_HI+1,x
+		rts
+		.pend
+		
 neg_f		.proc
 		; -- push -flt back on stack
 		jsr  pop_float_fac1
@@ -364,24 +384,63 @@ neg_f		.proc
 		jsr  c64.NEGOP
 		jmp  push_fac1_as_result
 		.pend
-
 		
-add_w		.proc
-		; -- push word+word
-		.warn "addw check correctness"
-		inx
-		clc
-		lda  ESTACK_LO,x
-		adc  ESTACK_LO+1,x
+inv_word	.proc
+		lda  ESTACK_LO+1,x
+		eor  #255
 		sta  ESTACK_LO+1,x
-		lda  ESTACK_HI,x
-		adc  ESTACK_HI+1,x
+		lda  ESTACK_HI+1,x
+		eor  #255
 		sta  ESTACK_HI+1,x
 		rts
 		.pend
+
+not_byte	.proc
+		lda  ESTACK_LO+1,x
+		beq  +
+		lda  #0
+		beq ++
++		lda  #1
++		sta  ESTACK_LO+1,x
+		rts
+		.pend
 		
-add_uw		.proc
-		.warn "add_uw check correctness"
+not_word	.proc
+		lda  ESTACK_LO + 1,x
+		ora  ESTACK_HI + 1,x
+		beq  +
+		lda  #0
+		beq  ++
++		lda  #1
++		sta  ESTACK_LO + 1,x
+		sta  ESTACK_HI + 1,x
+		rts
+		.pend
+		
+abs_b		.proc
+		; -- push abs(byte) on stack (as byte)
+		lda  ESTACK_LO+1,x
+		bmi  neg_b
+		rts
+		.pend
+		
+abs_w		.proc
+		; -- push abs(word) on stack (as word)
+		lda  ESTACK_HI+1,x
+		bmi  neg_w
+		rts
+		.pend
+
+abs_f		.proc
+		; -- push abs(float) on stack (as float)
+		jsr  pop_float_fac1
+		stx  SCRATCH_ZPREGX
+		jsr  c64.ABS
+		jmp  push_fac1_as_result
+		.pend
+
+add_w		.proc
+		; -- push word+word / uword+uword
 		inx
 		clc
 		lda  ESTACK_LO,x
@@ -394,13 +453,16 @@ add_uw		.proc
 		.pend
 		
 sub_w		.proc
-		rts	; @todo inline?
-		.warn "sub_w not implemented"
-		.pend
-
-sub_uw		.proc
-		rts	; @todo inline?
-		.warn "sub_uw not implemented"
+		; -- push word-word
+		inx
+		sec
+		lda  ESTACK_LO+1,x
+		sbc  ESTACK_LO,x
+		sta  ESTACK_LO+1,x
+		lda  ESTACK_HI+1,x
+		sbc  ESTACK_HI,x
+		sta  ESTACK_HI+1,x
+		rts
 		.pend
 
 mul_b		.proc
@@ -445,7 +507,7 @@ div_uw		.proc
 
 remainder_b	.proc
 		rts
-		.warn "remainder_b not implemented"
+		.warn "remainder_b via div_b?"
 		.pend
 		
 remainder_ub	.proc
@@ -463,12 +525,12 @@ remainder_ub	.proc
 		
 remainder_w	.proc
 		rts
-		.warn "remainder_w not implemented"
+		.warn "remainder_w not implemented - via div_w"
 		.pend
 		
 remainder_uw	.proc
 		rts
-		.warn "remainder_uw not implemented"
+		.warn "remainder_uw not implemented - via div_uw"
 		.pend
 		
 equal_w		.proc
@@ -790,11 +852,6 @@ func_cos	.proc
 		jmp  push_fac1_as_result
 		.pend
 		
-func_abs	.proc
-		rts
-		.warn "abs not implemented"
-		.pend
-		
 func_tan	.proc
 		; -- push tan(f) back onto stack
 		jsr  pop_float_fac1