Re: [wsjt-devel] Question about QSO Start & End Times

[Black Michael via wsjt-devel]

I"m not running FT8 so questiondo you see the "Next" buttons getting 
enabled when progressing on auto sequence?
I get the feeling they are not being set which means it would skip setting the 
QSO start time...and is why you set start=end since start never gets set.
de Mike W9MDB

  From: John Nelson 
 To: Black Michael ; WSJT software development 
 
 Sent: Sunday, July 2, 2017 2:42 PM
 Subject: Re: [wsjt-devel] Question about QSO Start & End Times
   
Mike,

Like Neil I also notice that start time = end time in wsjtx.log and 
wsjtx_log.adi.  Running r on Mac OSX 10.11.

Extract from ALL.TXT

192430  0  0.6 1057 ~  CQ A92AA LL56        
192430 -16  0.4  945 ~  EK6RSC YO2NAA KN05  
192430  -9  0.6 1351 ~  MM0LGS CT1GVN IM67  
1924  Transmitting 14.079 MHz  FT8:  A92AA G4KLA IO92      
192500  -2  0.3 1055 ~  CQ A92AA LL56        
192500 -13  0.1 1210 ~  UR4UM YO2NAA KN05    
192500  2  0.2 1431 ~  MM0LGS CT1GVN R-05  
1925  Transmitting 14.079 MHz  FT8:  A92AA G4KLA IO92      
192530  0  0.8 1057 ~  G4KLA A92AA -17      
192530  5  0.2 1430 ~  MM0LGS CT1GVN 73    
1925  Transmitting 14.079 MHz  FT8:  A92AA G4KLA R+00      
192600  1  0.3 1056 ~  G4KLA A92AA -17      
1926  Transmitting 14.079 MHz  FT8:  A92AA G4KLA R+01      
192630  2  0.3 1056 ~  G4KLA A92AA -17      
192630 -17  0.2 1675 ~  AK1P CT1GVN IM67    
1926  Transmitting 14.079 MHz  FT8:  A92AA G4KLA R+02      
192700  1  0.3 1055 ~  G4KLA A92AA RRR      
192700 -13  0.2 1674 ~  AK1P CT1GVN IM67    
1927  Transmitting 14.079 MHz  FT8:  A92AA G4KLA 73        
192730  2  0.3 1055 ~  G4KLA A92AA 73      

Extract from wsjtx.log

2017-07-02,19:27,2017-07-02,19:27,A92AA,LL56,14.080055,FT8,+02,-17,20W,v1.7.1-dev
 r 10.11,Fawaz

Extract from wsjtx_log.adi

A92AA LL56 FT8 +02 -17 
20170702 1927 20170702 1927 
20m 14.080055 G4KLA IO92ak 
20W v1.7.1-dev r 10.11 Fawaz 

— John G4KLA

   --
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] JTSDK Usage and Policy Guidelines

[Eric Gruff]

Thanks, Greg. I think we all agree that we don't want to see the beta
updates no longer made available.

May I suggest to all the group users that if we see folks requesting install
packages, we offer to help them set up the JTSDK environment on their
machines, in the "teach a man to fish" spirit. They can then easily build
the releases by typing the few magic phrases "build-hamlib3" and
"build-wsjtx package". If they aren't willing, then they can wait for the
official release. Those that are really motivated to get on FT8 will have to
put out the relatively minimal effort.

I had a few minor issues because I didn't follow directions, then started
the install again, and it was very easy. I am happy to help if anyone needs
it. I know some of you are Linux and Mac expert users, so I think we will
have it covered. 

73,

Eric NC6K

-Original Message-
From: Greg Beam [mailto:ki7m...@gmail.com] 
Sent: Sunday, July 2, 2017 3:47 PM
To: 'WSJT software development' ; 'WSJT
Group' 
Subject: [wsjt-devel] JTSDK Usage and Policy Guidelines

Hello All,

Appropriate Use
==

JTSDK exists to help users build and "test" Joe's software, not distribute
it.  We all want these fantastic applications, but, we "users" must respect
the developers requests.

I've received several reports of WSJTX packages built with JTSDK being
distributed by end-users. If this continues, I'll have no choice but to
remove the package build feature from JTSDK. Joe has, on many occasions,
requested that package installers not be re-distributed by anyone other than
the development team. Please respect their wishes!


For "Non-Developers" / "Testers" - (Setup && Usage)
===
For those of you "testing" Joe's software, a few things to note:

* DO NOT distribute the packages you build for yourself! < nuff said >
* Make sure, "before posting bugs", you are running the latest SVN revision
* Setup JTSDK-QT as follows:

- Open JTSDK-QT [1]
- Set the following options; can be on one line separated by a " ; "
semicolon,
or you can set them individually. If individually, you do not need the " ; "
:

enable-separate ; disable-quiet ; disable-skipsvn ; enable-autosvn ;
enable-clean ; enable-rcfg ; enable-qt55

You can, if desired, enable-autorun, which will launch the newly built
version for you.

* Check your options are set properly with [2]:   list-options
* Close, then reopen JTSDK-QT
* Build Hamlib3:  build-hamlib3
* Build the Release Install target (unless otherwise directed by a developer
for troubleshooting): 

build-wsjtx rinstall

You can build WSJT-X as often as you like, and should do; especially when
new features have been added, or the development team is requesting broader
use testing.

* Build an installable package ( DO NO DISTRIBUTE THESE < nuff said again >
):
* For most users, this is "not needed, nor desired for testing"

build-wsjtx package


Problems with JTSDK
=

If you have problems with JTSDK (not WSJT-X), email me off line, and we'll
get it sorted out.

@ ki...@yahoo.com
@ ki7m...@gmail.com

Or, post a bug on GitHub: 

Linux: https://github.com/KI7MT/jtsdk-nix/issues
Windows: https://github.com/KI7MT/jtsdk-win/issues


Documentation
=

For Windows, documentation is available online at:
http://jtsdk-win.readthedocs.io/en/latest/

For Linux:   /usr/share/doc/jtsdk/

If you find bugs in the Docs, report them to the appropriate GitHub location
above.


73's
Greg, KI7MT

[1] main-menu.png
[2] options.png




--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] JTSDK Usage and Policy Guidelines

[Greg Beam]

Hello All,

Appropriate Use
==

JTSDK exists to help users build and "test" Joe's software, not distribute
it.  We all want these fantastic applications, but, we "users" must respect
the developers requests.

I've received several reports of WSJTX packages built with JTSDK being
distributed by end-users. If this continues, I'll have no choice but to
remove the package build feature from JTSDK. Joe has, on many occasions,
requested that package installers not be re-distributed by anyone other than
the development team. Please respect their wishes!


For "Non-Developers" / "Testers" - (Setup && Usage)
===
For those of you "testing" Joe's software, a few things to note:

* DO NOT distribute the packages you build for yourself! < nuff said >
* Make sure, "before posting bugs", you are running the latest SVN revision
* Setup JTSDK-QT as follows:

- Open JTSDK-QT [1]
- Set the following options; can be on one line separated by a " ; "
semicolon, 
or you can set them individually. If individually, you do not need the " ; "
:

enable-separate ; disable-quiet ; disable-skipsvn ; enable-autosvn ;
enable-clean ; enable-rcfg ; enable-qt55

You can, if desired, enable-autorun, which will launch the newly built
version for you.

* Check your options are set properly with [2]:   list-options
* Close, then reopen JTSDK-QT
* Build Hamlib3:  build-hamlib3
* Build the Release Install target (unless otherwise directed by a developer
for troubleshooting): 

build-wsjtx rinstall

You can build WSJT-X as often as you like, and should do; especially when
new features have been added, or the development team is requesting broader
use testing.

* Build an installable package ( DO NO DISTRIBUTE THESE < nuff said again >
):
* For most users, this is "not needed, nor desired for testing"

build-wsjtx package


Problems with JTSDK
=

If you have problems with JTSDK (not WSJT-X), email me off line, and we'll
get it sorted out.

@ ki...@yahoo.com
@ ki7m...@gmail.com

Or, post a bug on GitHub: 

Linux: https://github.com/KI7MT/jtsdk-nix/issues
Windows: https://github.com/KI7MT/jtsdk-win/issues


Documentation
=

For Windows, documentation is available online at:
http://jtsdk-win.readthedocs.io/en/latest/

For Linux:   /usr/share/doc/jtsdk/

If you find bugs in the Docs, report them to the appropriate GitHub location
above.


73's
Greg, KI7MT

[1] main-menu.png
[2] options.png

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Thanks John,

> I get the same as Joe for the decoding of Tamás’s files.  Running r and 
> Mac OSX 10.11.6

Here’s a summary of what I’ve found so far.

I have three machines. Two run OS X and one is a linux box. One of the Macs and 
the linux box show the smaller DT values that others are seeing.

On the machine that shows the large DTs, if I decode using jt9 from the command 
line, I do not see the large DT values. 

The machines on which the DTs look OK are built with Qt 5.6 (Mac) and 5.5 
(Ubuntu).

The Mac Mini (on which the DTs are large is built with Qt 5.7.

So far, this seems to suggest that this is a Qt issue. Does this make sense?

Steve k9an



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Error building on OS X

[Tim Carlson]

Hi Everyone,

I’m grateful to Bill (G4WJS) for helping me resolve the issue.  I’m posting the 
solution in case anyone else makes the same mistake as I did.

The issue was that I downloaded the latest release candidate of CMake, rather 
than the current release version.  After installing 3.8.2 (rather than 
3.9.0-rc5), everything complied without issues.

-Tim (KD0GYG)

> On Jul 1, 2017, at 9:41 PM, Tim Carlson  wrote:
> 
> Also, here’s my configuration results:
> 
> $ FC=gfortran-mp-5 cmake -D 
> CMAKE_PREFIX_PATH="~/Qt3/5.9.1/clang_64;~/hamlib-prefix;/opt/local" -D 
> CMAKE_INSTALL_PREFIX=~/wsjtx-prefix -D 
> CMAKE_OSX_SYSROOT=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.11.sdk
>  ~/wsjtx-prefix/src
> -- Building wsjtx-1.7.1-devel
> -- **
> -- Building for for: Darwin-x86_64
> -- **
> -- Boost version: 1.63.0
> -- Could NOT find OpenMP_C (missing: OpenMP_C_FLAGS OpenMP_C_LIB_NAMES) 
> (found version "1.0")
> -- Could NOT find OpenMP_CXX (missing: OpenMP_CXX_FLAGS OpenMP_CXX_LIB_NAMES) 
> (found version "1.0")
> -- Checking for module 'hamlib'
> --   Found hamlib, version 3.2~git
> -- Found hamlib 
> -- hamlib_INCLUDE_DIRS: 
> /Users/tim/hamlib-prefix/include;/opt/local/include/libusb-1.0
> -- hamlib_LIBRARIES: 
> -L/Users/tim/hamlib-prefix/lib;-L/opt/local/lib;-lhamlib;-lusb-1.0;-lobjc;-Wl,-framework,IOKit;-Wl,-framework,CoreFoundation
> -- hamlib_LIBRARY_DIRS: /Users/tim/hamlib-prefix/lib;/opt/local/lib
> -- Asking qmake for QT_PLUGINS_DIR and got 
> /Users/tim/Qt3/5.9.1/clang_64/plugins
> -- Asking qmake for QT_IMPORTS_DIR and got 
> /Users/tim/Qt3/5.9.1/clang_64/imports
> -- Asking qmake for QT_DATA_DIR and got /Users/tim/Qt3/5.9.1/clang_64
> -- Configuring done
> -- Generating done
> -- Build files have been written to: /Users/tim/wsjtx-prefix/build
> 
> I just noticed the “Could NOT find OpenMP_C" and OpenMP_CXX lines - does that 
> have anything to do with the subsequent error?
> 
> -Tim
> 
>> On Jul 1, 2017, at 9:20 PM, Bill Somerville > > wrote:
>> 
>> On 02/07/2017 03:49, Tim Carlson wrote:
>>> Thanks, Bill.  That did cause it to recompile everything, but it still 
>>> quits at the same point.
>>> 
>> Hi Tim,
>> 
>> I think the problem is that the compiler is missing. Have you recently 
>> updated your MacPorts? If so then the compiler name has probably changed. 
>> You will need to do the cmake configure stage again in that case. The 
>> compiler may well be called gfortran-mp-6 now if you have the latest version.
>> 
>> 73
>> Bill
>> G4WJS.
>> 
>> 
>> --
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org ! 
>> http://sdm.link/slashdot 
>> ___
>> wsjt-devel mailing list
>> wsjt-devel@lists.sourceforge.net 
>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
> 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] PSKReporter displayed freq for FT8

[Borja Marcos]

> On 2 jul 2017, at 20:34, Lloyd Kirk  wrote:
> 
> I noticed that pskreporter is displaying the exact RX freq for jt65/9 signals 
> but not for FT8. Is this by design?
> 
> It seems to be displaying the frequency that is configured in the 
> settings->frequencies for that mode in WSJT-X.
> 
> Was just curious. Using r currently.

I’ve got a data feed of the pskreporter data (I’m exploring some options to add 
visualizations, currently feeding it to an Elasticsearch cluster) and the data 
I am receiving for FT8 only includes frequencies rounded to the KHz. It would 
be a bit odd for the pskreporter server to round the frequencies. Are you sure 
the versions of WSJT-X being used are reporting it properly?

From what I’ve seen, pskreporter doesn’t seem to correct frequencies in any 
way. At least for other modes, there’s no reason for it to include provisions 
for a mode that does not yet exist “publicly”.





Borja - EA2EKH



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Question about QSO Start & End Times

[John Nelson]

Mike,

Like Neil I also notice that start time = end time in wsjtx.log and 
wsjtx_log.adi.   Running r on Mac OSX 10.11.

Extract from ALL.TXT

192430   0  0.6 1057 ~  CQ A92AA LL56
192430 -16  0.4  945 ~  EK6RSC YO2NAA KN05   
192430  -9  0.6 1351 ~  MM0LGS CT1GVN IM67   
1924  Transmitting 14.079 MHz  FT8:  A92AA G4KLA IO92  
192500  -2  0.3 1055 ~  CQ A92AA LL56
192500 -13  0.1 1210 ~  UR4UM YO2NAA KN05
192500   2  0.2 1431 ~  MM0LGS CT1GVN R-05   
1925  Transmitting 14.079 MHz  FT8:  A92AA G4KLA IO92  
192530   0  0.8 1057 ~  G4KLA A92AA -17  
192530   5  0.2 1430 ~  MM0LGS CT1GVN 73 
1925  Transmitting 14.079 MHz  FT8:  A92AA G4KLA R+00  
192600   1  0.3 1056 ~  G4KLA A92AA -17  
1926  Transmitting 14.079 MHz  FT8:  A92AA G4KLA R+01  
192630   2  0.3 1056 ~  G4KLA A92AA -17  
192630 -17  0.2 1675 ~  AK1P CT1GVN IM67 
1926  Transmitting 14.079 MHz  FT8:  A92AA G4KLA R+02  
192700   1  0.3 1055 ~  G4KLA A92AA RRR  
192700 -13  0.2 1674 ~  AK1P CT1GVN IM67 
1927  Transmitting 14.079 MHz  FT8:  A92AA G4KLA 73
192730   2  0.3 1055 ~  G4KLA A92AA 73   

Extract from wsjtx.log

2017-07-02,19:27,2017-07-02,19:27,A92AA,LL56,14.080055,FT8,+02,-17,20W,v1.7.1-dev
 r 10.11,Fawaz

Extract from wsjtx_log.adi

A92AA LL56 FT8 +02 -17 
20170702 1927 20170702 1927 
20m 14.080055 G4KLA IO92ak 
20W v1.7.1-dev r 10.11 Fawaz 

— John G4KLA
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] PSKReporter displayed freq for FT8

[Lloyd Kirk]

I noticed that pskreporter is displaying the exact RX freq for jt65/9 
signals but not for FT8. Is this by design?


It seems to be displaying the frequency that is configured in the 
settings->frequencies for that mode in WSJT-X.


Was just curious. Using r currently.

Lloyd Kirk
WB5HUP

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[John Nelson]

Steve,

I get the same as Joe for the decoding of Tamás’s files.  Running r and Mac 
OSX 10.11.6

— John G4KLA



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Colors - FT8

[Charles Suckling]

Hi Joe

Seems to be working fine, learning how to use it now.

Charlie

-Original Message-
From: Joe Taylor [mailto:j...@princeton.edu] 
Sent: 02 July 2017 19:01
To: WSJT software development
Subject: Re: [wsjt-devel] Colors - FT8

Hi Kari,

> The 'worked before' feature is also not working with FT8.
> 
> I think this could be fixed by adding test for
> FT8 on line 105 of "logbook/adif.cpp"

Thanks!  I believe it's fixed in r.

-- Joe, K1JT


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Colors - FT8

[Joe Taylor]

Hi Kari,


The 'worked before' feature is also not working with FT8.

I think this could be fixed by adding test for
FT8 on line 105 of "logbook/adif.cpp"


Thanks!  I believe it's fixed in r.

-- Joe, K1JT

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Minimum transmission length for FT8

[Charles Suckling]

tnx

-Original Message-
From: Joe Taylor [mailto:j...@princeton.edu] 
Sent: 02 July 2017 18:45
To: WSJT software development
Subject: Re: [wsjt-devel] Minimum transmission length for FT8

Hi Charlie,

> What fraction of the transmission is needed for a successful decode 
> given 'good' S/N?
> 
> I've seen some "short" messages decode just now (using r7776).

The code rate is 1/2, so without AP decoding the bare minimum must be 
about half a transmission.  I've seen done good decodes with not much 
more than that.
-- Joe


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Minimum transmission length for FT8

[Joe Taylor]

Hi Charlie,

What fraction of the transmission is needed for a successful decode 
given ‘good’ S/N?


I’ve seen some “short” messages decode just now (using r7776).


The code rate is 1/2, so without AP decoding the bare minimum must be 
about half a transmission.  I've seen done good decodes with not much 
more than that.

-- Joe

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] UI Suggestion (not bug!)

[George J Molnar]

Is it possible to change the “Disable Tx after sending 73” function to “Disable 
Tx after logging QSO”?

With it checked, we get one 73, which is sometimes not enough. Usually it is. 
But whenever a QSO is logged, we are -for sure- done with Tx5 and can go on 
from there, either with a CQ or stop.

7776 looking pretty good from here. Many thanks.

(Does anyone need any .wav files still?)


George J Molnar
Nevada, USA
KF2T  @GJMolnar





--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] Minimum transmission length for FT8

[Charles Suckling]

Hi Joe/Steve

 

What fraction of the transmission is needed for a successful decode given
'good' S/N?

 

I've seen some "short" messages decode just now (using r7776).

 

73

 

Charlie

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] FT8 "AFC"??

[Dave 'Doc' Corio]

Does the new FT8 have any sort of built in AFC compensation? Every 
so often I see a station slowly creeping up the band while in QSO. It 
looks very much like the way PSK acts when both stations have their AFC 
control checked and they "chase" each other up or down the band.


It will only happen to one station at a time that I've seen, so it 
doesn't appear to be my audio that is slowly shifting, although I have 
seen it once during a QSO I was having.


The shift is only a few Hz at a time, but even in the short length 
of an FT8 QSO, it adds up to 15 Hz or so.


Tnx es 73
Dave - KB3MOW

--



---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Colors - FT8

[Karza]

On 07/02/2017 06:14 PM, James Shaver (N2ADV) wrote:

Running r7775 (Win 7) I've noticed and others have brought to my attention that 
the colors for new DXCC and new call aren't showing up in the RX window.  CQ in 
Message and My Call in message are working.

73,

Jim S.
N2ADV


The 'worked before' feature is also not working with FT8.

I think this could be fixed by adding test for
FT8 on line 105 of "logbook/adif.cpp"

73's de Kari
oh2gqc

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7776 Split problem

[Bill Somerville]

Hi Dan,

you've got it now, hi hi.

73
Bill
G4WJS.

On 02/07/2017 18:02, Dan Malcolm wrote:


Bill,

At the moment it’s 1636.  Perhaps the light is dawning. Would split 
mode only shift the Tx frequency in order to keep the audio frequency 
in bounds, and not to shift if Tx frequency is already in bounds?


*From:*Bill Somerville [mailto:g4...@classdesign.com]
*Sent:* Sunday, July 02, 2017 11:33 AM
*To:* wsjt-devel@lists.sourceforge.net
*Subject:* Re: [wsjt-devel] r7776 Split problem

On 02/07/2017 17:29, Dan Malcolm wrote:

I think I understand, but perhaps I didn’t communicate well. 
Prior to r7776 the B BFO did the Tx (and A BFO does the Rx) and

shifted frequency to 14.078.500 as I expected, in compliance with
the recommend split mode setting.  Now, Tx still occurs on the B
VFO, but does not shift to 14.078.500 as it did previously.  The B
VFO remains at 14.079.000 during Tx.  This seems to be a change in
Tx split mode behavior.  I understand the DF frequency will change
depending on the Tx DF, but I thought the TX shift using split
mode was supposed to keep audio frequencies in 1200-2000Hz range
and to do that required a 500Hz downshift during Tx.

Hi Dan,

what is your Tx DF when you observe the Tx dial frequency (VFO B) at 
14.079 ?


73
Bill
G4WJS.



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7776 Split problem

[Dan Malcolm]

I just tested my last reply to you, and it appears I understand "split mode"
better.  Thanks.

 

From: Bill Somerville [mailto:g4...@classdesign.com] 
Sent: Sunday, July 02, 2017 11:33 AM
To: wsjt-devel@lists.sourceforge.net
Subject: Re: [wsjt-devel] r7776 Split problem

 

On 02/07/2017 17:29, Dan Malcolm wrote:

I think I understand, but perhaps I didn't communicate well.  Prior to r7776
the B BFO did the Tx (and A BFO does the Rx) and shifted frequency to
14.078.500 as I expected, in compliance with the recommend split mode
setting.  Now, Tx still occurs on the B VFO, but does not shift to
14.078.500 as it did previously.  The B VFO remains at 14.079.000 during Tx.
This seems to be a change in Tx split mode behavior.  I understand the DF
frequency will change depending on the Tx DF, but I thought the TX shift
using split mode was supposed to keep audio frequencies in 1200-2000Hz range
and to do that required a 500Hz downshift during Tx.

Hi Dan,

what is your Tx DF when you observe the Tx dial frequency (VFO B) at 14.079
?

73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7776 Split problem

[Dan Malcolm]

Bill,

At the moment it's 1636.  Perhaps the light is dawning.  Would split mode
only shift the Tx frequency in order to keep the audio frequency in bounds,
and not to shift if Tx frequency is already in bounds?

 

From: Bill Somerville [mailto:g4...@classdesign.com] 
Sent: Sunday, July 02, 2017 11:33 AM
To: wsjt-devel@lists.sourceforge.net
Subject: Re: [wsjt-devel] r7776 Split problem

 

On 02/07/2017 17:29, Dan Malcolm wrote:

I think I understand, but perhaps I didn't communicate well.  Prior to r7776
the B BFO did the Tx (and A BFO does the Rx) and shifted frequency to
14.078.500 as I expected, in compliance with the recommend split mode
setting.  Now, Tx still occurs on the B VFO, but does not shift to
14.078.500 as it did previously.  The B VFO remains at 14.079.000 during Tx.
This seems to be a change in Tx split mode behavior.  I understand the DF
frequency will change depending on the Tx DF, but I thought the TX shift
using split mode was supposed to keep audio frequencies in 1200-2000Hz range
and to do that required a 500Hz downshift during Tx.

Hi Dan,

what is your Tx DF when you observe the Tx dial frequency (VFO B) at 14.079
?

73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7776 Split problem

[Bill Somerville]

On 02/07/2017 17:29, Dan Malcolm wrote:
I think I understand, but perhaps I didn’t communicate well. Prior to 
r7776 the B BFO did the Tx (and A BFO does the Rx) and shifted 
frequency to 14.078.500 as I expected, in compliance with the 
recommend split mode setting.  Now, Tx still occurs on the B VFO, but 
does not shift to 14.078.500 as it did previously.  The B VFO remains 
at 14.079.000 during Tx.  This seems to be a change in Tx split mode 
behavior.  I understand the DF frequency will change depending on the 
Tx DF, but I thought the TX shift using split mode was supposed to 
keep audio frequencies in 1200-2000Hz range and to do that required a 
500Hz downshift during Tx.


Hi Dan,

what is your Tx DF when you observe the Tx dial frequency (VFO B) at 
14.079 ?


73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7776 Split problem

[Dan Malcolm]

Bill,

I think I understand, but perhaps I didn't communicate well.  Prior to r7776
the B BFO did the Tx (and A BFO does the Rx) and shifted frequency to
14.078.500 as I expected, in compliance with the recommend split mode
setting.  Now, Tx still occurs on the B VFO, but does not shift to
14.078.500 as it did previously.  The B VFO remains at 14.079.000 during Tx.
This seems to be a change in Tx split mode behavior.  I understand the DF
frequency will change depending on the Tx DF, but I thought the TX shift
using split mode was supposed to keep audio frequencies in 1200-2000Hz range
and to do that required a 500Hz downshift during Tx.

 

From: Bill Somerville [mailto:g4...@classdesign.com] 
Sent: Sunday, July 02, 2017 11:17 AM
To: wsjt-devel@lists.sourceforge.net
Subject: Re: [wsjt-devel] r7776 Split problem

 

On 02/07/2017 17:13, Dan Malcolm wrote:

I just compiled r7776 and found that there is a "split" problem.  In
Settings under the Radio tab I have enabled Rig for Split Operation.  I am
running Kenwood TS-2000 for the Rig model.  That setting has been working
well until I compiled the new WSJT-X 1.7.1-devel r7776.  Prior to this Tx
occurred on the B CFO at 14.78.500Mhz.  Now it is still using the B VFO but
no offset occurs.  I may have confused things a bit because I also
recompiled HamLIb3.

 

Dan Malcolm

K4SHQ

Hi Dan,

the Tx VFO dial frequency is not fixed, it depends on the Tx DF you select.
If you control-click around the waterfall you should see that each 500Hz
section uses a different Tx VFO dial frequency. That is assuming you are
using one of the multi-decode modes JT9, JT65 or FT8.

73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Tamás Fábián]

Dear Steve and everyone,

r7776 decodes the files as Joe described previously, and seems to
successfully decode current on-air signals.

(I'm using Ubuntu Linux 17.04 "zesty".)

Cheers,

Tamas HA5FTL

2017-07-02 18:02 GMT+02:00 Steven Franke :

> Ah - I should have RTFM. I am not used to using the FFTW library with
> real-valued input vectors. Thanks!
> Steve
>
> > On Jul 2, 2017, at 10:54 AM, Bill Somerville 
> wrote:
> >
> > On 02/07/2017 16:47, Steven Franke wrote:
> >> So why did you need to add the extra element at the end of cx?
> >
> > The real input generates a complex DFT with one extra element
> sizeof(input)/2 + 1.
> >
> > http://www.fftw.org/fftw3_doc/Real_002ddata-DFT-Array-
> Format.html#Real_002ddata-DFT-Array-Format
> >
> > 73
> > Bill
> > G4WJS.
> >
> >
> > 
> --
> > Check out the vibrant tech community on one of the world's most
> > engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> > ___
> > wsjt-devel mailing list
> > wsjt-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
>
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Now that Bill has fixed the memory issue, it would be interesting to see if we 
get the same DT values from Tamas’ files when running on Windows and OS X. The 
decodes that I get from r7776 on my OS X machine are the same as I had before:

130800  -7  0.9 1097 ~  EI4KF ES6DO KO27
- 6m
130815 -12  0.5 1098 ~  ES6DO EI4KF -13
- 6m
130830  -7  0.9 1097 ~  EI4KF ES6DO R-08
- 6m
130900 -17  0.9 1097 ~  EI4KF ES6DO 73
- 6m
131000 -10  0.9  857 ~  CQ I4WQH JN54
131000 -13  1.0  917 ~  CQ ES6DO KO27

Steve

> On Jul 2, 2017, at 3:54 PM, Bill Somerville  wrote:
> 
> On 02/07/2017 16:47, Steven Franke wrote:
>> So why did you need to add the extra element at the end of cx?
> 
> The real input generates a complex DFT with one extra element sizeof(input)/2 
> + 1.
> 
> http://www.fftw.org/fftw3_doc/Real_002ddata-DFT-Array-Format.html#Real_002ddata-DFT-Array-Format
> 
> 73
> Bill
> G4WJS.
> 
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7776 Split problem

[Bill Somerville]

On 02/07/2017 17:13, Dan Malcolm wrote:


I just compiled r7776 and found that there is a “split” problem.  In 
Settings under the Radio tab I have enabled Rig for Split Operation.  
I am running Kenwood TS-2000 for the Rig model.  That setting has been 
working well until I compiled the new WSJT-X 1.7.1-devel r7776.  Prior 
to this Tx occurred on the B CFO at 14.78.500Mhz.  Now it is still 
using the B VFO but no offset occurs.  I may have confused things a 
bit because I also recompiled HamLIb3.


Dan Malcolm

K4SHQ


Hi Dan,

the Tx VFO dial frequency is not fixed, it depends on the Tx DF you 
select. If you control-click around the waterfall you should see that 
each 500Hz section uses a different Tx VFO dial frequency. That is 
assuming you are using one of the multi-decode modes JT9, JT65 or FT8.


73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] r7776 Split problem

[Dan Malcolm]

I just compiled r7776 and found that there is a "split" problem.  In
Settings under the Radio tab I have enabled Rig for Split Operation.  I am
running Kenwood TS-2000 for the Rig model.  That setting has been working
well until I compiled the new WSJT-X 1.7.1-devel r7776.  Prior to this Tx
occurred on the B CFO at 14.78.500Mhz.  Now it is still using the B VFO but
no offset occurs.  I may have confused things a bit because I also
recompiled HamLIb3.

 

Dan Malcolm

K4SHQ

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Ah - I should have RTFM. I am not used to using the FFTW library with 
real-valued input vectors. Thanks!
Steve

> On Jul 2, 2017, at 10:54 AM, Bill Somerville  wrote:
> 
> On 02/07/2017 16:47, Steven Franke wrote:
>> So why did you need to add the extra element at the end of cx?
> 
> The real input generates a complex DFT with one extra element sizeof(input)/2 
> + 1.
> 
> http://www.fftw.org/fftw3_doc/Real_002ddata-DFT-Array-Format.html#Real_002ddata-DFT-Array-Format
> 
> 73
> Bill
> G4WJS.
> 
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Bill Somerville]

On 02/07/2017 16:47, Steven Franke wrote:

So why did you need to add the extra element at the end of cx?


The real input generates a complex DFT with one extra element 
sizeof(input)/2 + 1.


http://www.fftw.org/fftw3_doc/Real_002ddata-DFT-Array-Format.html#Real_002ddata-DFT-Array-Format

73
Bill
G4WJS.


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Bill - 
Thank you for finding this! 

But I still don’t understand what was wrong with the way that the cx array was 
sized.

Surely the size of the original x array (NMAX real elements) is equivalent to 
the size of the original cx array 2*(NMAX/2) complex elements...

I just did this test (using the original code):

   call four2a(x,NMAX,1,-1,0) !r2c FFT to freq domain
205 write(*,*) x(1:4)
206 write(*,*) cx(0:1)
207 write(*,*) x(NMAX-3:NMAX)
208 write(*,*) cx(NMAX/2-2:NMAX/2-1)

with these results:

  -21077.   0.   32318.1348  -31433.4023
 ( -21077.,  0.) (  32318.1348, -31433.4023)
   36066.5547   41465.5625   34216.6055   26828.1055
 (  36066.5547,  41465.5625) (  34216.6055,  26828.1055)

which seem to confirm that elements x(1:4) line up with cx(0:1) and that 
x(NMAX-3:NMAX) line up with cx(NMAX/2-2:NMAX/2-1), as expected. So why did you 
need to add the extra element at the end of cx?

Steve


> On Jul 2, 2017, at 10:12 AM, Bill Somerville  wrote:
> 
> On 02/07/2017 16:06, Bill Somerville wrote:
>> Steve, I didn't fully analyse the issue. Does the c1 vector in 
>> ft8_downsample() need to be one longer as well?
> 
> Hi Steve,
> 
> ok, I see not as it is complex to complex for the return to time domain.
> 
> 73
> Bill
> G4WJS.
> 
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] Colors - FT8

[James Shaver (N2ADV)]

Running r7775 (Win 7) I've noticed and others have brought to my attention that 
the colors for new DXCC and new call aren't showing up in the RX window.  CQ in 
Message and My Call in message are working.  

73,

Jim S. 
N2ADV

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Bill Somerville]

On 02/07/2017 16:06, Bill Somerville wrote:
Steve, I didn't fully analyse the issue. Does the c1 vector in 
ft8_downsample() need to be one longer as well?


Hi Steve,

ok, I see not as it is complex to complex for the return to time domain.

73
Bill
G4WJS.


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Bill Somerville]

On 02/07/2017 15:41, Bill Somerville wrote:

On 02/07/2017 14:28, Tamás Fábián wrote:

Dear Steve,

I get no decodes with FT8 and r7775.

I tried to call CQ and I got an answer, but could not decode that 
either. (The CQ was heard by multiple stations according to 
pskreporter.)


Five sample recordings are in a .zip under the following url: 
https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1


Cheers,

Tamas HA5FTL


Hi All,

it looks like r7773 breaks the FT8 decoder.

Steve, this is bad news as it is as I suspected and that change only 
moved the defect rather than fixing it.


We still have a stack corruption issue IMHO. I am observing this 
non-decoding on Windows so it is not Mac specific either.


73
Bill
G4WJS. 


Hi All,

this should be fixed now.

Steve, I didn't fully analyse the issue. Does the c1 vector in 
ft8_downsample() need to be one longer as well?


73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Bill Somerville]

On 02/07/2017 14:28, Tamás Fábián wrote:

Dear Steve,

I get no decodes with FT8 and r7775.

I tried to call CQ and I got an answer, but could not decode that 
either. (The CQ was heard by multiple stations according to pskreporter.)


Five sample recordings are in a .zip under the following url: 
https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1


Cheers,

Tamas HA5FTL


Hi All,

it looks like r7773 breaks the FT8 decoder.

Steve, this is bad news as it is as I suspected and that change only 
moved the defect rather than fixing it.


We still have a stack corruption issue IMHO. I am observing this 
non-decoding on Windows so it is not Mac specific either.


73
Bill
G4WJS.


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Erik Icket]

Hi Tamas, 

I decoded the same 6 decodes as Steve with build 7774 on Windows.

Hope this helps, 

ON4PB,
Erik.

--

Message: 3
Date: Sun, 02 Jul 2017 13:37:21 +
From: Steven Franke 
To: Joe Taylor 
Subject: Re: [wsjt-devel] r7775 FT8 decoding
Message-ID: 
Content-Type: text/plain; charset=utf-8

Hi Tamas,

Your files produce the following decodes on r7775 here:

- 6m
130800  -7  0.9 1097 ~  EI4KF ES6DO KO27
- 6m
130815 -12  0.5 1098 ~  ES6DO EI4KF -13
- 6m
130830  -7  0.9 1097 ~  EI4KF ES6DO R-08
- 6m
130900 -17  0.9 1097 ~  EI4KF ES6DO 73
- 6m
131000 -10  0.9  857 ~  CQ I4WQH JN54
131000 -13  1.0  917 ~  CQ ES6DO KO27

As far as I can tell, all of the signals that are visible on the waterfall
are decoded. I am running on OS X here. Perhaps others can try to decode
your files to see if there is a problem on Windows.

Steve

> On Jul 2, 2017, at 1:28 PM, Tam?s F?bi?n  wrote:
> 
> Dear Steve,
> 
> I get no decodes with FT8 and r7775.
> 
> I tried to call CQ and I got an answer, but could not decode that 
> either. (The CQ was heard by multiple stations according to 
> pskreporter.)
> 
> Five sample recordings are in a .zip under the following url: 
> https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=
> 1
> 
> Cheers,
> 
> Tamas HA5FTL
> 
> 
> 2017-07-02 13:58 GMT+02:00 Steven Franke :
> For future reference, I?ve attached a plot summarizing the performance of
the FT8 decoder in r7775. The 50% decoding threshold is at about -21 dB.
These results pertain to an AWGN channel (no fading) and were obtained using
.wav files generated by ft8sim. Each .wav file contains 25 signals. A total
of 40 such files (1000 potential decodes) was analyzed at each SNR. The
ft8sim DT parameter was set to 0.5s for this simulation.
> 
> I should note that the idealized AWGN channel simulation used for this
benchmark does not reflect the type of conditions that we typically see in
real life on HF or 6m. This is mainly intended as a performance benchmark
for development purposes - not an advertisement of actual performance in the
real world!
> 
> Steve k9an
> 
> 
> 


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Another observation. Watching you CQ on 20m now, I am seeing your DT jump 
between 0.1 and 0.6…. 

> On Jul 2, 2017, at 2:07 PM, Steven Franke  wrote:
> 
> Hi Joe - 
> The DTs that you got from Tamas’ files on Windows are different than what I 
> got on OS X. Did you change anything that could explain this? If not - this 
> could be connected with the problem that I was having yesterday...
> Steve
> 
>> On Jul 2, 2017, at 2:03 PM, Joe Taylor  wrote:
>> 
>> Hi Tamas,
>> 
>> Your files decode OK here in WSJT-X r7775 (running on Windows).
>> 
>>  20m
>> 130800  -7  0.6 1097 ~  EI4KF ES6DO KO27
>>  20m
>> 130815 -12  0.2 1098 ~  ES6DO EI4KF -13
>>  20m
>> 130830  -7  0.6 1097 ~  EI4KF ES6DO R-08
>>  20m
>> 130900 -17  0.7 1097 ~  EI4KF ES6DO 73
>>  20m
>> 131000 -11  0.6  857 ~  CQ I4WQH JN54 ~Italy
>> 131000 -13  0.7  917 ~  CQ ES6DO KO27 !Estonia
>> 
>>  -- 73, Joe, K1JT
>> 
>> On 7/2/2017 9:28 AM, Tamás Fábián wrote:
>>> Dear Steve,
>>> I get no decodes with FT8 and r7775.
>>> I tried to call CQ and I got an answer, but could not decode that either. 
>>> (The CQ was heard by multiple stations according to pskreporter.)
>>> Five sample recordings are in a .zip under the following url: 
>>> https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1
>>> Cheers,
>>> Tamas HA5FTL
>>> 2017-07-02 13:58 GMT+02:00 Steven Franke >> >:
>>>   For future reference, I’ve attached a plot summarizing the
>>>   performance of the FT8 decoder in r7775. The 50% decoding threshold
>>>   is at about -21 dB. These results pertain to an AWGN channel (no
>>>   fading) and were obtained using .wav files generated by ft8sim. Each
>>>   .wav file contains 25 signals. A total of 40 such files (1000
>>>   potential decodes) was analyzed at each SNR. The ft8sim DT parameter
>>>   was set to 0.5s for this simulation.
>>>   I should note that the idealized AWGN channel simulation used for
>>>   this benchmark does not reflect the type of conditions that we
>>>   typically see in real life on HF or 6m. This is mainly intended as a
>>>   performance benchmark for development purposes - not an
>>>   advertisement of actual performance in the real world!
>>>   Steve k9an
>>>   
>>> --
>>>   Check out the vibrant tech community on one of the world's most
>>>   engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>>   ___
>>>   wsjt-devel mailing list
>>>   wsjt-devel@lists.sourceforge.net
>>>   
>>>   https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>>>   
>>> --
>>> Check out the vibrant tech community on one of the world's most
>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>> ___
>>> wsjt-devel mailing list
>>> wsjt-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>> 
>> --
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> ___
>> wsjt-devel mailing list
>> wsjt-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
> 
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Hi Joe - 
The DTs that you got from Tamas’ files on Windows are different than what I got 
on OS X. Did you change anything that could explain this? If not - this could 
be connected with the problem that I was having yesterday...
Steve

> On Jul 2, 2017, at 2:03 PM, Joe Taylor  wrote:
> 
> Hi Tamas,
> 
> Your files decode OK here in WSJT-X r7775 (running on Windows).
> 
>  20m
> 130800  -7  0.6 1097 ~  EI4KF ES6DO KO27
>  20m
> 130815 -12  0.2 1098 ~  ES6DO EI4KF -13
>  20m
> 130830  -7  0.6 1097 ~  EI4KF ES6DO R-08
>  20m
> 130900 -17  0.7 1097 ~  EI4KF ES6DO 73
>  20m
> 131000 -11  0.6  857 ~  CQ I4WQH JN54 ~Italy
> 131000 -13  0.7  917 ~  CQ ES6DO KO27 !Estonia
> 
>   -- 73, Joe, K1JT
> 
> On 7/2/2017 9:28 AM, Tamás Fábián wrote:
>> Dear Steve,
>> I get no decodes with FT8 and r7775.
>> I tried to call CQ and I got an answer, but could not decode that either. 
>> (The CQ was heard by multiple stations according to pskreporter.)
>> Five sample recordings are in a .zip under the following url: 
>> https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1
>> Cheers,
>> Tamas HA5FTL
>> 2017-07-02 13:58 GMT+02:00 Steven Franke > >:
>>For future reference, I’ve attached a plot summarizing the
>>performance of the FT8 decoder in r7775. The 50% decoding threshold
>>is at about -21 dB. These results pertain to an AWGN channel (no
>>fading) and were obtained using .wav files generated by ft8sim. Each
>>.wav file contains 25 signals. A total of 40 such files (1000
>>potential decodes) was analyzed at each SNR. The ft8sim DT parameter
>>was set to 0.5s for this simulation.
>>I should note that the idealized AWGN channel simulation used for
>>this benchmark does not reflect the type of conditions that we
>>typically see in real life on HF or 6m. This is mainly intended as a
>>performance benchmark for development purposes - not an
>>advertisement of actual performance in the real world!
>>Steve k9an
>>
>> --
>>Check out the vibrant tech community on one of the world's most
>>engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>___
>>wsjt-devel mailing list
>>wsjt-devel@lists.sourceforge.net
>>
>>https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>>
>> --
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> ___
>> wsjt-devel mailing list
>> wsjt-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Joe Taylor]

Hi Tamas,

Your files decode OK here in WSJT-X r7775 (running on Windows).

 20m
130800  -7  0.6 1097 ~  EI4KF ES6DO KO27
 20m
130815 -12  0.2 1098 ~  ES6DO EI4KF -13
 20m
130830  -7  0.6 1097 ~  EI4KF ES6DO R-08
 20m
130900 -17  0.7 1097 ~  EI4KF ES6DO 73
 20m
131000 -11  0.6  857 ~  CQ I4WQH JN54 ~Italy
131000 -13  0.7  917 ~  CQ ES6DO KO27 !Estonia

-- 73, Joe, K1JT

On 7/2/2017 9:28 AM, Tamás Fábián wrote:

Dear Steve,

I get no decodes with FT8 and r7775.

I tried to call CQ and I got an answer, but could not decode that 
either. (The CQ was heard by multiple stations according to pskreporter.)


Five sample recordings are in a .zip under the following url: 
https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1


Cheers,

Tamas HA5FTL


2017-07-02 13:58 GMT+02:00 Steven Franke >:


For future reference, I’ve attached a plot summarizing the
performance of the FT8 decoder in r7775. The 50% decoding threshold
is at about -21 dB. These results pertain to an AWGN channel (no
fading) and were obtained using .wav files generated by ft8sim. Each
.wav file contains 25 signals. A total of 40 such files (1000
potential decodes) was analyzed at each SNR. The ft8sim DT parameter
was set to 0.5s for this simulation.

I should note that the idealized AWGN channel simulation used for
this benchmark does not reflect the type of conditions that we
typically see in real life on HF or 6m. This is mainly intended as a
performance benchmark for development purposes - not an
advertisement of actual performance in the real world!

Steve k9an





--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/wsjt-devel





--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot



___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Steven Franke]

Hi Tamas,

Your files produce the following decodes on r7775 here:

- 6m
130800  -7  0.9 1097 ~  EI4KF ES6DO KO27
- 6m
130815 -12  0.5 1098 ~  ES6DO EI4KF -13
- 6m
130830  -7  0.9 1097 ~  EI4KF ES6DO R-08
- 6m
130900 -17  0.9 1097 ~  EI4KF ES6DO 73
- 6m
131000 -10  0.9  857 ~  CQ I4WQH JN54
131000 -13  1.0  917 ~  CQ ES6DO KO27

As far as I can tell, all of the signals that are visible on the waterfall are 
decoded. I am running on OS X here. Perhaps others can try to decode your files 
to see if there is a problem on Windows.

Steve

> On Jul 2, 2017, at 1:28 PM, Tamás Fábián  wrote:
> 
> Dear Steve,
> 
> I get no decodes with FT8 and r7775.
> 
> I tried to call CQ and I got an answer, but could not decode that either. 
> (The CQ was heard by multiple stations according to pskreporter.)
> 
> Five sample recordings are in a .zip under the following url: 
> https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1
> 
> Cheers,
> 
> Tamas HA5FTL
> 
> 
> 2017-07-02 13:58 GMT+02:00 Steven Franke :
> For future reference, I’ve attached a plot summarizing the performance of the 
> FT8 decoder in r7775. The 50% decoding threshold is at about -21 dB. These 
> results pertain to an AWGN channel (no fading) and were obtained using .wav 
> files generated by ft8sim. Each .wav file contains 25 signals. A total of 40 
> such files (1000 potential decodes) was analyzed at each SNR. The ft8sim DT 
> parameter was set to 0.5s for this simulation.
> 
> I should note that the idealized AWGN channel simulation used for this 
> benchmark does not reflect the type of conditions that we typically see in 
> real life on HF or 6m. This is mainly intended as a performance benchmark for 
> development purposes - not an advertisement of actual performance in the real 
> world!
> 
> Steve k9an
> 
> 
> 
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
> 
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! 
> http://sdm.link/slashdot___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] r7775 FT8 decoding

[Tamás Fábián]

Dear Steve,

I get no decodes with FT8 and r7775.

I tried to call CQ and I got an answer, but could not decode that either.
(The CQ was heard by multiple stations according to pskreporter.)

Five sample recordings are in a .zip under the following url:
https://www.dropbox.com/s/sgcfehcdwfrp7th/wsjtx-r7775-nodecode.zip?dl=1

Cheers,

Tamas HA5FTL


2017-07-02 13:58 GMT+02:00 Steven Franke :

> For future reference, I’ve attached a plot summarizing the performance of
> the FT8 decoder in r7775. The 50% decoding threshold is at about -21 dB.
> These results pertain to an AWGN channel (no fading) and were obtained
> using .wav files generated by ft8sim. Each .wav file contains 25 signals. A
> total of 40 such files (1000 potential decodes) was analyzed at each SNR.
> The ft8sim DT parameter was set to 0.5s for this simulation.
>
> I should note that the idealized AWGN channel simulation used for this
> benchmark does not reflect the type of conditions that we typically see in
> real life on HF or 6m. This is mainly intended as a performance benchmark
> for development purposes - not an advertisement of actual performance in
> the real world!
>
> Steve k9an
>
>
>
>
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] FT8 - build 7774 - A few GUI suggestions

[Erik Icket]

Hi,

 

Can I make a few very ** minor ** GUI suggestions ?

 

1.  Shouldn't we rename the TX even/1st label in JT8 mode to something
that explicitly refers to the quarter/kwadrant inside the minute into which
we will transmit ?

Something like "Start at 0 or 30 secs" ?

 

2.  In the main window, could we rename the "RX Frequency" window into
"RX/TX Activity" or "RX/TX Frequency" ? Is there any reason why "TX" cannot
appear at the top of the right list window ?

 

3.  Would it be difficult to improve the "Copy to Clipboard" for a
single line ? For example : click one line, the line highlights, and a ctrl
C copies it to the clipboard. A doublie click would still copy the line to
the RX Frequency window and kick off the transmit (if appropriate).

This would be very helpful when copying over a frame into an external
logbook.

 

Thanks for considering this and best 73's from Belgium,

 

Erik

ON4PB

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] r7775 FT8 decoding

[Steven Franke]

For future reference, I’ve attached a plot summarizing the performance of the 
FT8 decoder in r7775. The 50% decoding threshold is at about -21 dB. These 
results pertain to an AWGN channel (no fading) and were obtained using .wav 
files generated by ft8sim. Each .wav file contains 25 signals. A total of 40 
such files (1000 potential decodes) was analyzed at each SNR. The ft8sim DT 
parameter was set to 0.5s for this simulation.

I should note that the idealized AWGN channel simulation used for this 
benchmark does not reflect the type of conditions that we typically see in real 
life on HF or 6m. This is mainly intended as a performance benchmark for 
development purposes - not an advertisement of actual performance in the real 
world! 

Steve k9an


fig_ft8.pdf
Description: Adobe PDF document


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Error building on OS X

[Bill Somerville]

On 02/07/2017 04:41, Tim Carlson wrote:

Also, here’s my configuration results:

$ FC=gfortran-mp-5 cmake -D
CMAKE_PREFIX_PATH="~/Qt3/5.9.1/clang_64;~/hamlib-prefix;/opt/local"
-D CMAKE_INSTALL_PREFIX=~/wsjtx-prefix -D

CMAKE_OSX_SYSROOT=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.11.sdk
~/wsjtx-prefix/src
-- Building wsjtx-1.7.1-devel
-- **
-- Building for for: Darwin-x86_64
-- **
-- Boost version: 1.63.0
-- Could NOT find OpenMP_C (missing: OpenMP_C_FLAGS
OpenMP_C_LIB_NAMES) (found version "1.0")
-- Could NOT find OpenMP_CXX (missing: OpenMP_CXX_FLAGS
OpenMP_CXX_LIB_NAMES) (found version "1.0")
-- Checking for module 'hamlib'
--   Found hamlib, version 3.2~git
-- Found hamlib
-- hamlib_INCLUDE_DIRS:
/Users/tim/hamlib-prefix/include;/opt/local/include/libusb-1.0
-- hamlib_LIBRARIES:

-L/Users/tim/hamlib-prefix/lib;-L/opt/local/lib;-lhamlib;-lusb-1.0;-lobjc;-Wl,-framework,IOKit;-Wl,-framework,CoreFoundation
-- hamlib_LIBRARY_DIRS: /Users/tim/hamlib-prefix/lib;/opt/local/lib
-- Asking qmake for QT_PLUGINS_DIR and got
/Users/tim/Qt3/5.9.1/clang_64/plugins
-- Asking qmake for QT_IMPORTS_DIR and got
/Users/tim/Qt3/5.9.1/clang_64/imports
-- Asking qmake for QT_DATA_DIR and got /Users/tim/Qt3/5.9.1/clang_64
-- Configuring done
-- Generating done
-- Build files have been written to: /Users/tim/wsjtx-prefix/build


I just noticed the “Could NOT find OpenMP_C" and OpenMP_CXX lines - 
does that have anything to do with the subsequent error?


-Tim


Hi Tim,

those OpenMP messages are expected because we use the Apple Clang 
compilers for C and C++ which do not support OpenMP but I would expect 
to see some messages about OpenMP and FORTRAN. There is definitely 
something odd going on with setting up the FORTRAN compiler. Did you 
delete the whole build tree before re-configuring? CMake will use cached 
configuration information if it is there and I suspect that there is 
something referenced in the CMake that has gone or moved.


73
Bill
G4WJS.

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel