v1.4.5 - fixes for negative numbers

• forth: / now uses floored signed division
• forth: bugfix for negative number parsing (2- was
  interpreted as -20)
• forth: made 2/ sign-extend negative numbers
• forth: added spaces source >in 2over 2swap m+ m*
sm/rem fm/mod dabs dnegate s>d
• forth: changed word order of double-cell integers
• vi: backspace on start of line now doesn't join lines
• doc: mention in tutorial that default is hexadecimal

Thanks to Christian Johansson and polluks for suggestions!

v1.4.4

• forth: added u. unloop leave cell+ cells page key?
• forth: allow input of negative numbers
• forth: optimized cmove, cmove>
• forth: made . print signed numbers
• forth: various bugfixes
• doc: simplified & improved intro + tutorial, minor improvements
• vi: bugfix, exiting editor disabled key repeat

Thanks to Christian Johansson and polluks for bug reports!

v1.4.3

• forth: made +loop work like it should
• forth: removed <=, >=
• forth: renamed <, > to u<, u>
• forth: renamed s<, s> to <, >
• forth: renamed d* to um*
• forth: optimized loop, r>, >r, r@, i
• forth: disabled tail call elimination for (
• vi: faster key delay+repeat
• vi: hid fg
• vi: various bugfixes
• doc: documented min, max, within

v1.4.2

• renamed EOL comment # to \ (PC) or £ (C64)
• renamed :asm => code
• renamed ;asm => ;code
• renamed tell => type
• disabled tail call elimination for s" : ]
• minor optimizations

v1.4.1 - tail call elimination

2015-08-13: v1.4.1

• forth: added tail call elimination, i.e., the practice of replacing
  jsr/rts with jmp. words that should not be subject to tail call
  elimination must be tagged with "no-tce".
• forth: renamed jsr => sys
• forth: renamed immed => immediate
• forth: added postpone
• forth: removed [compile]
• forth: inline drop, 2drop

v1.4 - subroutine threading

2015-08-08: v1.4

• forth: durexForth is now subroutine threaded instead
  of direct threaded. This gives a huge speed improvement,
  at the cost of increased RAM usage.
• forth: added "compile," word.
• disk: improved EOF handling.
• vi: bugfixed CTRL+u/d.
• vi: bugfix: "fg" was by mistake hidden.
• doc: added inline assembly example.
• doc: bugfixed turtle graphics demo.

v1.33

• vi: changed half-page scrolling from U/D to CTRL+u/d
• vi: many bugfixes and optimizations
• forth: many small optimizations
• forth: made cmove> ANS Forth compliant
• doc: updated memory map

v1.32 - dual parameter stack

2015-03-27: v1.32

• switched to dual parameter stack that is split up in MSB/LSB sections.
  this is a nice optimization!
• optimized "branch"
• renamed "var" to "variable"
• added "depth" (gives depth of parameter stack)
• added "sp0" (bottom address of LSB parameter stack)
• added "sp1" (bottom address of MSB parameter stack)
• removed "sp@"
• bugfixed & optimized "ls"
• moved zptmp, zptmp2, zptmp3 and ip to new locations

(On left side: v1.31. On right side: v1.32. Shorter distance between raster bars indicates that v1.32 gives a nice speedup.)

v1.31

• changed ' ['] to be according to standard
• re-add sid.fs, which contains SID manipulation words
• bugfix . in mml player

v1.3

• Music Macro Language (MML) support!
• Renamed some assembly mnemonics.
  asla, lsra, rora, rola, =>
  asl,a lsr,a ror,a rol,a
• Allow strings longer than 256 bytes.
• Optimized jsr. (Thanks to Kevin Lee Reno!)