diff --git a/mandel.s b/mandel.s
index db072be..b7c323b 100644
--- a/mandel.s
+++ b/mandel.s
@@ -34,7 +34,9 @@ count_frames    = $be ; u8
 ; free space $bf
 
 count_iters     = $c0 ; u16
-; free space c2-cb
+text_col        = $c2 ; u8
+text_row        = $c3 ; u8
+; free space c4-cb
 temp            = $cc ; u16
 temp2           = $ce ; u16
 
@@ -140,16 +142,16 @@ KEY_0     = 50
 
 strings:
 str_self:
-    .byte "MANDEL-6502"
+    .byte "MANDEL-6502", 0
 str_self_end:
 str_speed:
-    .byte "us/iter: "
+    .byte "us/iter: ", 0
 str_speed_end:
 str_run:
-    .byte " RUN"
+    .byte " RUN", 0
 str_run_end:
 str_done:
-    .byte "DONE"
+    .byte "DONE", 0
 str_done_end:
 
 str_self_len = str_self_end - str_self
@@ -1241,57 +1243,50 @@ done:
     rts
 .endproc
 
-.macro draw_text_indirect col, len, strptr
-    ; clobbers A, X
-    .local loop
-    .local done
-    .local padding
-    ldx #0
+; in/out: column in text_col
+; in: row in text_row @fixme implement
+; in: pointer to string in INBUFF
+; clobbers x/y/a/temp
+.proc draw_string
+    drawptr = temp
+    strptr = INBUFF
+
+    clc
+    lda #.lobyte(textbuffer)
+    adc text_col
+    sta temp
+    lda #.hibyte(textbuffer)
+    adc #0
+    sta temp + 1
+
+    ldy #0
 loop:
-    cpx #len
-    beq done
-    txa
-    tay
     lda (strptr),y
-    pha ; save the char for terminator check
-    and #$7f ; strip the high bit (terminator)
-    tay
-    lda char_map,y
-    sta textbuffer + col,x
-    inx
+    ; if char's null, terminate c-style
+    beq done
+    ; save the char for terminator check
+    pha
+    ; strip the high bit (terminator)
+    and #$7f
+    tax
+    lda char_map,x
+    sta (drawptr),y
+    iny
 
     pla
-    bmi padding
+    ; _last_ char has high bit set in atari rom routines
+    bmi done
     jmp loop
 
-padding:
-    ldy #32 ; space
-    lda char_map,y
-
-    cpx #len
-    beq done
-    sta textbuffer + col,x
-    inx
-    jmp padding
-
 done:
-.endmacro
+    ; move the text column pointer
+    tya
+    clc
+    adc text_col
+    sta text_col
 
-.macro draw_text col, len, cstr
-    ; clobbers A, X
-    .local loop
-    .local done
-    ldx #0
-loop:
-    cpx #len
-    beq done
-    ldy cstr,x
-    lda char_map,y
-    sta textbuffer + col,x
-    inx
-    jmp loop
-done:
-.endmacro
+    rts
+.endproc
 
 .proc vblank_handler
     inc count_frames
@@ -1506,8 +1501,24 @@ zero_byte_loop:
 
 .proc status_bar
     ; Status bar
-    draw_text 0, str_self_len, str_self
-    draw_text 40 - str_run_len, str_run_len, str_run
+
+    lda #0
+    sta text_col
+    lda #0
+    sta text_row
+    lda #.lobyte(str_self)
+    sta INBUFF
+    lda #.hibyte(str_self)
+    sta INBUFF + 1
+    jsr draw_string
+
+    lda #(40 - str_run_len)
+    sta text_col
+    lda #.lobyte(str_run)
+    sta INBUFF
+    lda #.hibyte(str_run)
+    sta INBUFF + 1
+    jsr draw_string
 
     rts
 .endproc
@@ -1758,13 +1769,19 @@ update_status:
     clc
     jsr IFP
 
-    ; convert to ASCII in INBUFF
+    lda #speed_start
+    sta text_col
+    lda #0
+    sta text_row
+    lda #.lobyte(str_speed)
+    sta INBUFF
+    lda #.hibyte(str_speed)
+    sta INBUFF + 1
+    jsr draw_string
+
+    ; convert to ASCII in INBUFF and print
     jsr FASC
-
-    ; print the first 6 digits
-    draw_text speed_start, str_speed_len, str_speed
-    draw_text_indirect speed_start + str_speed_len, speed_precision, INBUFF
-
+    jsr draw_string
 skipped:
 
     ; sx += fill_level[fill_masks] + 1
@@ -1812,7 +1829,17 @@ fill_loop_done:
 
 loop:
     ; finished
-    draw_text 40 - str_done_len, str_done_len, str_done
+
+    lda #(40 - str_done_len)
+    sta text_col
+    lda #0
+    sta text_row
+    lda #.lobyte(str_done)
+    sta INBUFF
+    lda #.hibyte(str_done)
+    sta INBUFF + 1
+    jsr draw_string
+
     jsr keycheck
     beq loop
     jmp main_loop