History file for asm64 - the first multipass n64 assembler

All compliments or complaints ONLY to stage@ping.at !

05.10.97 - Released v.995 win32 -------------------------------------------------------------

** Don't be ashamed 'bout that mini-update. More to follow!

Fixed:  Branch encoding changed 27th time ;) and now seems to be right.
	  Constants can now be 32 bits wide. (Hopefully)
	  Both things pointed out by LEM (and others!), thx.
Changed:Behaviour of command-line parameters. Just to nerve you.
	  From now on only win32 (windoze95) relz, just to benefit from a few things.
	  (Hope there aren't too much dos users around. Sorry you guyzz.)
	  Btw, a hundred ppl promised me to write a front-end, but noone did...Thanx!

Still NOT fixed:
        Address range only 31bits!!!

	  Almost 5 month no update, must have been a timewarp... 

24.05.97 - Released v.992 DOS ---------------------------------------------------------------

Fixed:  Incorrect Base encoding in several commands. Thx to Flake/TRSI.

23.04.97 - Released v.991 DOS ---------------------------------------------------------------

Fixed:  Branches now encoded right. (2bits shifted to right!)
        Offset calculating in Base/Offset was often wrong! Should work now.

09.04.97 - Released v.99 DOS ----------------------------------------------------------------

New:    Pseudo-Ops .v64header, .romname, .romid, .romcountry, .manufacturer
        for producing target file with DoctorV64 header, but w/out CHECKSUM!
        If anybody knows 'bout the 64-bit checksum, please mail!
        +, -, *, /, | (boolean or), & (boolean and), (, ) included (thx ZEN!).
        Attention: Divison by zero causes runtime-error. Thats ok! ;)

Fixed   #include renamed to .include and seems ok!
	  Some bugs with label and constant names. Shorter names weren't allowed
	  if the string was a substring from another.
        Checking for duplicate symbols.
        .bigendian and .littleendian doesn't exist anymore...
	  I think there was a big misunderstanding. Output image is ALWAYS bigendian
        but byte swapping (within WORDS) can now turned on by .bytechange and
	  turned off by .nobytechange. (So I think for your Doc's you need to turn
        on .bytechange!) Please tell me if this still is not the truth...!

27.03.97 - Released V.971 DOS --------------------------------------------------------------

Fixed:  A few bugs in the parsing engine...

26.03.97 - Released v.97 DOS ---------------------------------------------------------------

Fixed:	All "j"umps should now work perfectly... (But please don't check with r4300i patch-
        worx 0.2 by Hartec or jovis' disassembler since on my opinion they're wrong.
	  Asgard's wizard64 indeed tells me that I'm right. My manual says: 26bit address
	  shifted 2bits left and combined........!)
	  ; (remark operator) bug fixed. 
        Bug in string management fixed!

25.03.97 - Released v.96 DOS ---------------------------------------------------------------

Fixed:  .wantsize got size/4  -  now ok!
New:	  .byte now accepts string in quotes

24.03.97 - Released v.95 DOS ---------------------------------------------------------------
