[hpsdr] Would the real DttSP please stand up?

[Frank Brickle]

Hi All --

Larry Gadallah, VE6VQ/W7 passed this along. I do read the hpsdr list
irregularly and so would have probably seen it in the fullness of time.

I personally have no idea what state the DttSP code used in PowerSDR might
be, nor have I for a long time. I do know that, despite claims to the
contrary, Flex has made changes and additions and failed to make them
available, in contravention of the GPL, but I have been too taken up with
other issues to pursue any remedy as yet.

The code in DttSP is not completely OS-independent, but it has been
compiled and executed on Linux and BSD without change. That was due in
large part to testing and debugging by AB2WF, who wrung out the code on OS
X.

There appears to be no reason for further work on DttSP. I stopped at the
point -- for example, in providing facilities for predistortion, in 2008 --
when anybody stopped trying to integrate any new features into an existing
application. I was also maintaining a parallel version that didn't use the
JACK audio subsystem for signal I/O. As far as I know there had been only
one user of that version and so I abandoned it around 2008 as well.

All of my subsequent efforts in this area have been in the development of
new system whose foundation was distributed processing of multirate
signals, with provisions for arbitrary subband extraction and insertion,
and time as well as frequency agility -- for example, the ability to
"scroll back and forth " a signal in time for a useful period on one or
more selected subchannels. Once again, the path that current RF hardware
development seems to have been taking makes the implementation of this
system needlessly complicated. But perhaps that's a discussion for another
time. The development of DttSP was largely a labor of love. That wasn't a
mistake but it's not something to repeat.

73
Frank
AB2KT/VE3


On Sun, May 11, 2014 at 7:22 PM, Warren C. Pratt <warren at wpratt.com> wrote:

> ***** High Performance Software Defined Radio Discussion List *****
>
>
>  Hi Larry,
>
> We have generally moved away from DttSP.  From your list below, it looks
> like the latest version we used in PowerSDR was:
>
> W5WC/PowerSDR_HPSDR_mRX/Source/DttSP         321d 19h   3049
>
> That sounds about right as I recall.
>
> As far as I know, Flex still uses DttSP in their version of PowerSDR.  Our
> openHPSDR evolution was to first re-write parts of that version and then
> later, as I mentioned, move on to  a newly developed library, "wdsp".
>
> With regard to your DttSP questions:  I believe that DttSP was written
> with the goal of being computer architecture and OS INdependent.  The
> version in PowerSDR has a fairly substantial interface piece that is more
> or less tied to the software architecture of PowerSDR.  The good news is
> that there's an interface there and it can be used in other applications;
> the bad news is that it limits the flexibility of the underlying library.
>
> I believe that there must be several users of DttSP out there.  openHPSDR
> has not made any effort to share fixes and improvements or participate in a
> general DttSP repository, at least for quite some time.  All the changes we
> made are, of course, open source and available to anyone to use subject to
> the terms of the GNU GPL v2 license.
>
> 73,
> Warren  NR0V
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help:
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>



-- 
Tell me, what is it you plan to do with your one wild and precious life? --
Mary Oliver
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20140512/2cf53773/attachment-0003.htm>