From: msmcdoug@no.spam.iinet   
      
   Tim Mann wrote:   
      
   > Yes.  Note that "bits 4 and 5" are defined using 0-origin big-ending   
   > counting, that the data bit pattern is 0xa1, and that the sync mark is   
   > preceded by a 00 with normal clocking and written 3 times.   
   >   
   > There's also another kind of sync pattern used before the index mark on   
   > disks that have one.  It's 0xc2 with missing clock between bits 3 and 4.   
   >   
   > Since you've already looked at jv2dmk.c, try looking at decoder.txt in   
   > the cw2dmk package for more info.   
      
   Shall do, thanks Tim! I should also thank you for the Model III MicroLabs   
   Hires documentation you put together, which I only discovered yesterday - an   
   invaluable reference when I implemented it on my Model III!   
      
   I'm currently trying to wrap my head around FM and MFM encoding, and how the   
   WD9216 can separate both types of encoding without knowing which it is...   
   but I'm sure I'll get there...   
      
   BTW It's fun getting back into the TRS-80 - whilst testing my floppy   
   implementation I've been playing the Big Five games again whilst waiting for   
   compiles to complete (I went scrounging for my old list of TRS-80 hi scores   
   as I did pretty well on Cosmic Fighter yesterday, but alas couldn't find   
   it)... and also discovered another great game I've never even seen before -   
   Nukliex, which I saw reviewed on trs80.org! Not to mention finding Meteor   
   Mission 1 which I'd always wondered about...   
      
   Regards,   
      
   --   
   |              Mark McDougall                | "Electrical Engineers do it   
   |     |   with less resistance!"   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)