Loading ...
Sorry, an error occurred while loading the content.

Re: gmfsk-hkj version 0.48 posted

Expand Messages
  • jhaynesatalumni
    ... Well, one of the joys of Linux is that if you don t like the way the precompiled program works you are free to get the source and compile it yourself. Not
    Message 1 of 9 , Apr 11 9:42 AM
    View Source
    • 0 Attachment
      --- In linuxham@yahoogroups.com, Hamish Moffatt <hamish@...> wrote:

      > BTW, for the purposes of distributions (like Debian), compile-time
      > options are no good...
      >
      Well, one of the joys of Linux is that if you don't like the way the
      precompiled program works you are free to get the source and compile
      it yourself. Not being a Real Programmer I can't say how hard it is
      to convert a compile-time option into a run-time option.
    • Kevin der Kinderen
      ... For something like gMFSK-hkj, I would compile it locally and then use something like checkinstall to create the .deb package and install it. I found this
      Message 2 of 9 , Apr 11 12:40 PM
      View Source
      • 0 Attachment
        On 4/11/06, jhaynesatalumni <jhhaynes@...> wrote:

        Well, one of the joys of Linux is that if you don't like the way the
        precompiled program works you are free to get the source and compile
        it yourself.  Not being a Real Programmer I can't say how hard it is
        to convert a compile-time option into a run-time option.
         
        For something like gMFSK-hkj, I would compile it locally and then use something like checkinstall to create the .deb package and install it. I found this helped to keep track of software where I wanted to tweak the configuration and still keep it in the package tree.
         
        Kev - K4VD

         
      • w1hkj
        ... Hamlib is a very nice general purpose one size fits all interface library. I have nothing against or for it. I tried using it from with gmfsk with limited
        Message 3 of 9 , Apr 11 3:11 PM
        View Source
        • 0 Attachment
          Hamish Moffatt wrote:
          On Tue, Apr 11, 2006 at 08:48:44AM -0400, w1hkj wrote:
            
          Hamish Moffatt wrote:
              
          BTW, for the purposes of distributions (like Debian), compile-time
          options are no good...
                
          I am open to suggestions to resolve that issue Hamish.  Some of the 
          patches (the user interface for example) probably should just be made a 
          part of the application.  Others, like the interface to the CAT programs 
          (ArgoV, DeltaII, Icom728 and Kachina) could be run-time switches.
              
          Well in the case of the CAT-type changes, wouldn't it be proper for
          those changes to go into hamlib rather than directly into gMFSK?
          
          Hamish
            
          Hamlib is a very nice general purpose one size fits all interface library.  I have nothing against or for it.  I tried using it from with gmfsk with limited success (it doesn't support all of the i/o I desire).  It does not purport to be a rig interface program, just a rig i/o program.  The DeltaII, ArgoV, Icom728 and especially the Kachina CAT programs are rig control programs.  In fact the Kachina program is the only Linux program for controlling the Kachina (which is a computer interface only rig ... no front panel what-so-ever).  Controlling the Kachina using hamlib did not work very well.  I tried it.  So I elected to use my own i/o code to the transceivers.  But the mods to gmfsk do not impact on the continued use of hamlib for those users with a different transceiver or need.  If I had been successful in using hamlib for the CAT programs I still would have had to provide a means of communications between the gmfsk program and the transceiver control program.   Only one program at a time should be attempting to send commands to the transceiver.   If you look at the TenTecCAT.c source code you will see that gmfsk and the CAT programs communicate using shared memory.  Very fast, very efficient.  The actual i/o to the transceiver is done in the CAT programs, which by the way can also run stand-a-lone without gmfsk.

          Dave (hkj)
        Your message has been successfully submitted and would be delivered to recipients shortly.