This was started one hour after a mayor disk-crash

95-11-18  18:40  Panning is set to middle when not a Amiga MOD.
                  When it's M.K. FLT4 FLT8 M!K! it's set to 80% - 20% ratio
                  and vice versa.
                 
95-12-08  15:21  Started reworking the code.
                  Plan:
                  To make it a lib... So when a user doesn't want Sound-
                  Blaster support , he doesn't need to use link it!
                 
95-12-17  03:51  Started reworking the code again.
                  I have to add support to screen sync timing.
                 
96-03-28  12:00  Didn't updated this file in a while.. woops..
                  I rewrote the whole mixing stuff... supports bi-loops and
                  stuff! Rewrote most of the code , rearranged structures ,
                  prepared the mixing for sound-effects.. made a disk driver.
                  Meaning a i can actually debug my mixing routine.. it saves
                  mixed data to disk(?).
                 
96-03-29  10:39  Fixed bugs in the XM player causing major headaches to
                  listeners. Rewrote XM_Setupvoice.
96-03-29  12:57  Found some bugs(?) in XM player again... Command 5 & 6
                  doesn't work at all as the MOD format. Remade the
                  XM_setupvoice again to make it look like a random generated
                  text , again. Sux... BUT hey! It works?
96-03-29  16:56  Bugs Bugs Bugs
                  In Spaceman.xm the patterns contained 0xfe as a note number?
                  Fixed by making everything over 97(keyoff) 0. Ft2 does alike.
                  Added more exceptions into the code making it worse...
96-03-30  13:40  Stupid me... "jbe" not "jb" , caused keyoff's to be ignored!
96-03-31  00:20  Removed Megaem-Check in GUS driver. I think it made some
                  computers hang. It ruined the first open file in my file
                  table... (?)
96-03-31  16:51  Remade the PAS driver to work with my new mixer device.
                  The sorry thing is that i have NO way of testing it =(.
                  Restarted work on the WSS driver.
96-03-31  17:32  Optimized Mixer by 1 cycle per sample. Doesn't sound like
                  much , but I can't really see how slow (or fast) it is now
                  when having a P100.. Totally removed the old Mono-screen
                  debug-driver , useless...
96-03-31  ??:??  Optimized Premix -> playbuff mover with 1 cycle per sample.
96-04-01  14:38  Found an instrument in liquidation.xm with loop flag on and
                  loopsize = 0 , And when loop is active I set the sample end
                  to loop end.. Coz that's the way it is...
                  Fixed so gus wouldn't play samples with <2 size in loop or
                  not.
96-04-04  03:54  Misstake.. Nosound was running at double speed!? FIXED!
96-04-11  21:38  Tried to solve some SB1.0 -> SB2.0 Problems... I dont'
                  know for sure before someone tries it. Fixed som checks
                  in the GUS-driver causing a "memory-play". Found a zero
                  length Vol envelope in a XM , didn't know FT2 allowed it!
                  I've put in some checks that disables loop if it's 0 long,
                  FT2 does... Added Support for 16bit sample playing in the
                  Software mixer.. This isn't a REAL 16bit sample I'm mixing
                  since it would require 16 Mb Voltable or a Mul and div for
                  every sample besides the ordinary. I simply take the high
                  byte of a word =).ReEnabled 16bit support in XM.
96-04-12  05:46  Got a report of a lot of clicking in a module - Tested -
                  and yes even though my be efforts avoiding clicks , there
                  they were. Carefully checked code. Still clicks.. Suddenly
                  NO clicks? I don't know what the problem could be.
                  I've scheduled a total rwrite of the GUS driver , AGAIN?
                  But meanwhile - Refrashed the GUS driver a lot , reenabled
                  DMA transfer , if it fails it simply does it the old fashion
                  way. Put the error messages in a separate obj , if you want
                  to you can save about 1 kb by not including error messages,
                  but who would do that? Refreshed complete nms system a bit.
                  Made Software mixer more indepentent , handles up to any
                  number of channels , though max is set to 64.
                  That means that When using a Software mixing card you
                  will have 32 module channels and 32 sfx channels!?
                  When using a Wavetablecard , 30 module channels will be
                  available + 32 SFX. GUSMAX+Pnp will ofcoz have all the 32
                  module channels. Started work with REAL IMM.
                  Minor changes in timer devices.
96-07-05  ??.??  Finally : The new Pmode/W is out with my requested functions
                  Support : Yes...
                  Fixed a nasty bug in SB causing clicks...
