[wsjt-devel] truncated decodes of 9LY1JM on fox mode

[Adrian Fabry]

Hello,

if the problem was already addressed I apologize.
I noticed that messages from non standard call 9LY1JM in fox mode are
truncated, but only when they are sent in separate carrier:
182330 -9 0.1 839 ~ F1UIJ 9LY1JM

182330 -9 0.1 899 ~ IZ2DVI 9LY1JM

182330 -9 0.1 959 ~ DL8MKG 9LY1JM

 40m

182400 -13 0.1 839 ~ F1UIJ RR73; IZ2DVI <9LY1JM> -08

182400 -12 0.1 899 ~ DL8MKG 9LY1JM

182400 -12 0.1 959 ~ SQ8N 9LY1JM +

 40m

182430 -10 0.1 839 ~ F1UIJ RR73; DL8MKG <9LY1JM> +02

182430 -9 0.1 899 ~ SQ8N 9LY1JM +

182430 -8 0.1 959 ~ LY2BAW 9LY1JM

Using WSJT-X 2.0.0

73 Ady YO2NAA
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] DXpedition mode - dupes ignored

[Adrian Fabry]

Hi all,

big thanks to WSJT developers for creating the FT8 DXpedition mode.

Recently, trying to work OJ0C who was using fox mode, I was hit by the
"ignore dupes" feature on the fox side.
For some reason, I haven't received my RR73 from the fox.
I continued to send my R-05 for a while and then restart the QSO from TX1,
but despite the strong signal, I never received an answer again.
>From my point of view, the QSO was not finished, I had no way to know if
OJ0C logged the contact.

I understand that if the fox restarts WSJT, the dupe is possible, but this
probably would not be done often, if everything works fine.

Is there any workaround for this problem?
I would propose a feature for the fox to force working particular dupes by
choice, or to have an "enable dupes" option.
At least, the fox should resend an RR73 after decoding a lot of TX1 from
the hound.

TNX, 73 Ady YO2NAA
--
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 - Removal of dupes

[Adrian Fabry]

I saw that Steve made a change in r7931:
"*Only print dupes if SNR is higher than what has already been printed."*


IMO dupes are useful in some cases:
1.is interesting to get 2 or 3 decodes of the same call doppler
shifted by airplane
reflection
2. some stations have side bands, it's good to decode them to be sure and
then let them know (in privat !)
3. if dupes will be removed, I can bet we'll get a lot of posts like "I can
see a nice trace on waterfall but no decode"

I'm aware that only some of the dupes will be removed by the change.
I would suggest to have an option to remove dupes or not, if not too
complicate to implement.


TNX, 73 Ady
YO2NAA
--
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] Suggestions for three extra bits in FT8

[Adrian Fabry]

Sorry please ingonre these 2 lines:

  P5DX DB0AAA -10
DB0AAA P5DX R-14 QRZ



On Tue, Jul 18, 2017 at 11:00 AM, Adrian Fabry  wrote:

> Hi all,
>
> with these 3 bits we can have 8 combinations, actually 7 usable, one is by
> default (let's say 000) for current messages types (standard of free).
>
> I would propose to use 1 combination (001) for DXpedition mode:
> - 001 when CQ or QRZ in DXpedition mode. This will instruct callers to
> spread in freq and to expect a shorter sequence.
> - 001 when DX is confirming the report from the caller. Will display a
> message like:
> *YO2NAA P5DX RRR QRZ*
>
> This message if decoded by YO2NAA will stop TX and prompt for LOG (if
> enabled).
>
> The full sequence:
>
> *=CQ P5DX PM29*  - contains 001  -  The* =* sign before CQ is displayed
> by WSJT to indicate DXped mode CQ. In addition can also be a different
> color)
>   *P5DX YO2NAA KN05*
> *YO2NAA P5DX -17*
> *  P5DX YO2NAA R-12*
> *YO2NAA P5DX RRR QRZ*- contains 001  (actually the message is YO2NAA
> P5DX RRR, but QRZ will be displayed by WSJT when detect 001)
>
>   P5DX DB0AAA -10
> DB0AAA P5DX R-14 QRZ
> etc
>
> If the caller used TX2 instead of TX1:
>
> *=CQ P5DX PM29  * - contains 001
>  *P5DX YO2NAA -12*
> *YO2NAA P5DX R-17*
> *  P5DX YO2NAA RRR*
> *YO2NAA P5DX 73 QRZ* - contains 001
>
>
> Other combinations of the 3 bits can be used for contest modes. Can be
> different type of contests, MSK144 (CQ WW VHF) style or different.
>
> 73 Ady,
> YO2NAA
>
>
>
>
> On Wed, Jul 12, 2017 at 1:28 AM, Tim Carlson  wrote:
>
>> Thanks, everyone for the discussion.
>>
>> The biggest takeaway for me is that I shouldn't use the wide graph to
>> determine the quality of the signal, which helps.
>>
>> I do have a waterfall display on my IC-7300, but it has nowhere near the
>> resolution of the wsjt-x wide graph so it’s not very helpful either.
>>
>> I’ll be quiet now!  :)
>>
>> -Tim KD0GYG
>>
>> > On Jul 11, 2017, at 1:54 PM, James Shaver (N2ADV) 
>> wrote:
>> >
>> > If you throw attenuation at the signal and the issue goes away, the
>> cause is most likely on the receive side.  It also helps to have a
>> panadapter available (something that isn't tied into the AF chain).  The
>> waterfall itself should not be used to determine signal quality.
>> >
>> > 73,
>> >
>> > Jim S.
>> > N2ADV
>> >
>> >> On Jul 11, 2017, at 2:45 PM, Bill Somerville 
>> wrote:
>> >>
>> >>> On 11/07/2017 19:12, Tim Carlson wrote:
>> >>> Would you consider that the first signal on the left is being
>> overdriven somewhat?  Or is that just a consequence of a stronger signal?
>> Or is it some atmospheric condition causing the signal to spread?
>> >>>
>> >> Hi Tim,
>> >>
>> >> hard to tell but it is all too common for the Tx audio level fed to
>> the rig to be too high. This causes non-linearity and will almost certainly
>> widen the transmitted signal. The main product of clipping is harmonics but
>> if the sender is using the split operating facility that is largely
>> innocuous due to the harmonics being above 3000 Hz and attenuated by the
>> rig's Tx SSB filter low pass cut off. Still there is no excuse for over
>> driving the audio input to the rig and there is a secondary consequence. At
>> each frequency shift of the modulation there is a minimal phase
>> discontinuity, we shift the phase without a glitch but nevertheless it is a
>> discontinuity. These small discontinuities fractionally widen the signal
>> when the audio is correctly matched through to the transmitter but over
>> drive widens them considerably. This is what you are seeing, a horizontal
>> spike on the waterfall at each tone shift.
>> >>
>> >> You must also be careful about attributing blame in these situations,
>> all of the above can happen on the receiving side too, so check your own
>> house is in order before accusing another of having a poor signal.
>> >>
>> >> 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
>> >&

Re: [wsjt-devel] Suggestions for three extra bits in FT8

[Adrian Fabry]

Hi all,

with these 3 bits we can have 8 combinations, actually 7 usable, one is by
default (let's say 000) for current messages types (standard of free).

I would propose to use 1 combination (001) for DXpedition mode:
- 001 when CQ or QRZ in DXpedition mode. This will instruct callers to
spread in freq and to expect a shorter sequence.
- 001 when DX is confirming the report from the caller. Will display a
message like:
*YO2NAA P5DX RRR QRZ*

This message if decoded by YO2NAA will stop TX and prompt for LOG (if
enabled).

The full sequence:

*=CQ P5DX PM29*  - contains 001  -  The* =* sign before CQ is displayed by
WSJT to indicate DXped mode CQ. In addition can also be a different color)
  *P5DX YO2NAA KN05*
*YO2NAA P5DX -17*
*  P5DX YO2NAA R-12*
*YO2NAA P5DX RRR QRZ*- contains 001  (actually the message is YO2NAA
P5DX RRR, but QRZ will be displayed by WSJT when detect 001)

  P5DX DB0AAA -10
DB0AAA P5DX R-14 QRZ
etc

If the caller used TX2 instead of TX1:

*=CQ P5DX PM29  * - contains 001
 *P5DX YO2NAA -12*
*YO2NAA P5DX R-17*
*  P5DX YO2NAA RRR*
*YO2NAA P5DX 73 QRZ* - contains 001


Other combinations of the 3 bits can be used for contest modes. Can be
different type of contests, MSK144 (CQ WW VHF) style or different.

73 Ady,
YO2NAA




On Wed, Jul 12, 2017 at 1:28 AM, Tim Carlson  wrote:

> Thanks, everyone for the discussion.
>
> The biggest takeaway for me is that I shouldn't use the wide graph to
> determine the quality of the signal, which helps.
>
> I do have a waterfall display on my IC-7300, but it has nowhere near the
> resolution of the wsjt-x wide graph so it’s not very helpful either.
>
> I’ll be quiet now!  :)
>
> -Tim KD0GYG
>
> > On Jul 11, 2017, at 1:54 PM, James Shaver (N2ADV) 
> wrote:
> >
> > If you throw attenuation at the signal and the issue goes away, the
> cause is most likely on the receive side.  It also helps to have a
> panadapter available (something that isn't tied into the AF chain).  The
> waterfall itself should not be used to determine signal quality.
> >
> > 73,
> >
> > Jim S.
> > N2ADV
> >
> >> On Jul 11, 2017, at 2:45 PM, Bill Somerville 
> wrote:
> >>
> >>> On 11/07/2017 19:12, Tim Carlson wrote:
> >>> Would you consider that the first signal on the left is being
> overdriven somewhat?  Or is that just a consequence of a stronger signal?
> Or is it some atmospheric condition causing the signal to spread?
> >>>
> >> Hi Tim,
> >>
> >> hard to tell but it is all too common for the Tx audio level fed to the
> rig to be too high. This causes non-linearity and will almost certainly
> widen the transmitted signal. The main product of clipping is harmonics but
> if the sender is using the split operating facility that is largely
> innocuous due to the harmonics being above 3000 Hz and attenuated by the
> rig's Tx SSB filter low pass cut off. Still there is no excuse for over
> driving the audio input to the rig and there is a secondary consequence. At
> each frequency shift of the modulation there is a minimal phase
> discontinuity, we shift the phase without a glitch but nevertheless it is a
> discontinuity. These small discontinuities fractionally widen the signal
> when the audio is correctly matched through to the transmitter but over
> drive widens them considerably. This is what you are seeing, a horizontal
> spike on the waterfall at each tone shift.
> >>
> >> You must also be careful about attributing blame in these situations,
> all of the above can happen on the receiving side too, so check your own
> house is in order before accusing another of having a poor signal.
> >>
> >> 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
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites

Re: [wsjt-devel] WSJT-X v1.8 User Guide

[Adrian Fabry]

Modes table corrected and updated.
If somebody can add the missing data, many thanks.

73 Ady
YO2NAA


On Fri, Jul 14, 2017 at 5:21 PM, Adrian Fabry  wrote:

> I'd kindly suggest to include a table with all the modes affected by
> the VHF/UHF/Microwave option
>  and what are the differences when VHF/UHF/Microwave is enabled / disabled.
>
> There is a warning on page 9:
> *"Enabling VHF/UHF/Microwave features necessarily disables the wideband
> multidecode*
> *capability of JT65. In most circumstances you should turn this feature
> off when*
> *operating at HF."*
>
> Are other modes, for example JT9A or FT8 affected by the VHF/UHF/Microwave
> option ?
>
>
> Another suggestion would be a merged Table 16.1, 16.2 and 16.3, to have
> all the modes to compare.
> I've attached an excel file.
>
>
> TNX, 73 Ady
> YO2NAA
>
> On Thu, Jul 13, 2017 at 12:05 AM, Joe Taylor  wrote:
>
>> Both links should work.  Hit Refresh.
>>
>> On 7/12/2017 4:56 PM, Morgen Benner wrote:
>>
>>> I think the link should have been:
>>>
>>> http://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjtx
>>> -main-1.8.0-rc1.html
>>>
>>> 73,
>>> KC9SWV - Morgen
>>> President, TARA
>>>
>>> On Wed, Jul 12, 2017 at 3:46 PM, Joe Taylor >> j...@princeton.edu>> wrote:
>>>
>>> Hi all,
>>>
>>> Most of the necessary updates of the WSJT-X User Guide to cover
>>> features in Version 1.8 are now in place.  Use *Help -> Online User
>>> Guide* or *F1* or click here:
>>>
>>> http://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjtx
>>> -main-1.7.1-devel.html
>>> <http://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjt
>>> x-main-1.7.1-devel.html>
>>>
>>> Some screen shots still need to be updated.  If you find errors or
>>> omissions, or anything else that could be improved, please report
>>> details here.
>>>
>>>  -- 73, 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
>>> <mailto:wsjt-devel@lists.sourceforge.net>
>>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>>> <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
>>
>
>


WSJT-X modes table.xlsx
Description: MS-Excel 2007 spreadsheet
--
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] increasing false decodes in Version R7895

[Adrian Fabry]

I confirm this in r7894, here are some wav files:

 

https://www.dropbox.com/s/2bmsv0trpn93qgt/FT8%20false.ZIP?dl=0

 

 

73 Ady YO2NAA

 

  _  

From: Bibinuk Dental [mailto:bibi...@hotmail.com] 
Sent: Sunday, July 16, 2017 12:30 PM
To: wsjt-devel@lists.sourceforge.net
Subject: [wsjt-devel] increasing false decodes in Version R7895

 

dear developer team 

my first bug report. pse advice if such report is welcome or not.

 

regards;


Remo es HB9SHD

 --- bug report  -

 

Program version: WSJT-X v.1.7.1-devel r7895 
Operating system:

- compiled and linked: Windows 10 Enterprise

- run: Windows 10 Pro 


Concise description of the problem

Since Version r7895 I experienced an increase of false decode, especially
below -20dB. I put all decodes in table below- good and false for comparison


Exact sequence of steps required to reproduce the problem

 

 

071430 -24  1.7 1077 ~  E23JSC S15NOS PO64 <-- false Decode

073030 -24 -3.1  333 ~  0R250L/X87PF2  <-- false Decode

073445 -24  1.8 1923 ~  -V/W+94-/?WXY  <-- false Decode

075630 -12  0.6 1152 ~  CQ EA7HG IM87  Spain

075700 -11  0.6 1152 ~  CQ EA7HG IM87  Spain

075700 -18  0.6 1128 ~  CQ EA7HG IM87  Spain

075700 -20  0.6 1176 ~  CQ EA7HG IM87  Spain

075730 -11  0.1 1152 ~  CQ EA7HG IM87  Spain

075730 -17  0.1 1127 ~  CQ EA7HG IM87  Spain

075800 -17  0.1 1151 ~  CQ EA7HG IM87  Spain

08 -11  0.1 1151 ~  CQ EA7HG IM87  Spain

08 -19  0.1 1126 ~  CQ EA7HG IM87  Spain

08 -19  0.2 1177 ~  CQ EA7HG IM87  Spain

080930 -24  1.9 2249 ~  NI8ZEH 5Q6 JQ97  <-- false Decode

081800 -24  0.8 2507 ~  AT1XDH SV0UUO RB02  <-- false Decode

082545 -24  1.0  903 ~  W /ZC14CEE3OA  <-- false Decode

083715 -24 -3.1  253 ~  Q24NUQ W43VPK PR23  <-- false Decode

084400 -20  0.8 1448 ~  SRI NO DECODE

085245 -24  1.8  233 ~  7E6GYCSK+37O7  <-- false Decode

085745 -19  0.8 1282 ~  CQ DL1IP JN49 ~Germany

090015 -20  0.2 1036 ~  CQ DH8IAT JN49 ~Germany

090345 -19  0.2 1012 ~  CQ DH8IAT JN49 ~Germany

091145 -24 -0.5 1678 ~  MX6FHH GE7ZKB BF13  <-- false Decode

091215 -18  0.7 1007 ~  CQ DH8IAT JN49 ~Germany

091245 -18  0.0 1008 ~  CQ DH8IAT JN49 ~Germany

091845 -17  0.8 1486 ~  CQ SV2JL KN10 ~Greece

091915 -15  0.8 1485 ~  CQ SV2JL KN10 ~Greece

 

--
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] MSK144 band activity first row

[Adrian Fabry]

Dear developers,

 

On MSK144 I think it would be helpful to have some info added to the first
row, similar to ISCAT mode (M N C T).

Please see attached screen captures.

 

Using here r 7873.

 

TNX, 73

Ady YO2NAA

 

--
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 small feature request - avoid TXing empty messages

[Adrian Fabry]

Hello,

 

Some times empty messages are transmited on FT8 (wrong TX5 maybe). (r7873)

I would propose FT8 to not TX them and show a warning.

 

 

TNX, 73 Ady YO2NAA

--
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] WSJT-X v1.8 User Guide

[Adrian Fabry]

I'd kindly suggest to include a table with all the modes affected by
the VHF/UHF/Microwave option
 and what are the differences when VHF/UHF/Microwave is enabled / disabled.

There is a warning on page 9:
*"Enabling VHF/UHF/Microwave features necessarily disables the wideband
multidecode*
*capability of JT65. In most circumstances you should turn this feature off
when*
*operating at HF."*

Are other modes, for example JT9A or FT8 affected by the VHF/UHF/Microwave
option ?


Another suggestion would be a merged Table 16.1, 16.2 and 16.3, to have all
the modes to compare.
I've attached an excel file.


TNX, 73 Ady
YO2NAA

On Thu, Jul 13, 2017 at 12:05 AM, Joe Taylor  wrote:

> Both links should work.  Hit Refresh.
>
> On 7/12/2017 4:56 PM, Morgen Benner wrote:
>
>> I think the link should have been:
>>
>> http://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjtx
>> -main-1.8.0-rc1.html
>>
>> 73,
>> KC9SWV - Morgen
>> President, TARA
>>
>> On Wed, Jul 12, 2017 at 3:46 PM, Joe Taylor > j...@princeton.edu>> wrote:
>>
>> Hi all,
>>
>> Most of the necessary updates of the WSJT-X User Guide to cover
>> features in Version 1.8 are now in place.  Use *Help -> Online User
>> Guide* or *F1* or click here:
>>
>> http://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjtx
>> -main-1.7.1-devel.html
>> > x-main-1.7.1-devel.html>
>>
>> Some screen shots still need to be updated.  If you find errors or
>> omissions, or anything else that could be improved, please report
>> details here.
>>
>>  -- 73, 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
>>
>>
> 
> --
> 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-X modes table.xlsx
Description: MS-Excel 2007 spreadsheet
--
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] WSJT-X 1.8 rc1 r7843 - FT8 weak signal decoded only on the RX freq

[Adrian Fabry]

TNX Bill, now it all makes sense.

 

73 Ady

 

  _  

From: Bill Somerville [mailto:g4...@classdesign.com] 
Sent: Thursday, July 13, 2017 7:18 PM
To: wsjt-devel@lists.sourceforge.net
Subject: Re: [wsjt-devel] WSJT-X 1.8 rc1 r7843 - FT8 weak signal decoded
only on the RX freq

 

On 13/07/2017 17:10, Adrian Fabry wrote:

using here WSJT-X 1.8.0.rc1 r7843 on Windows XP /32 SP3

 

On the attached FT8 wav file is a weak tropo 6m signal. It decodes -21 dB
but only when RX freq is placed on the signal (about 1012 Hz).

If I change the RX freq up or down for example to 1036 Hz I get no decode
anymore.

Is there anything that favors the RX freq when decoding? (beside being the
first decode in case of multiple signals )

 

On r7865 I built today,  it can't be decoded anymore regardless of the RX
freq. 

No complain, the signal is really on the edge, I just wanted to share this
interesting case.

HI Ady,

Steve and Joe are fine tuning FT8, there are trade offs between speed and
depth. We have used the Rx frequency offset as a constraint to enable a
deeper search. That may not remain the case or may even become a user
option. As smarter decoding algorithms are tried the speed/depth trade off
must be reassessed for optimum performance.

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] FT8 Issue

[Adrian Fabry]

Hi Ed,

 

How is your "Allow TX freq changes during transmision" ?

 

73 Ady YO2NAA

 

  _  

From: Ed Wilson via wsjt-devel [mailto:wsjt-devel@lists.sourceforge.net] 
Sent: Sunday, July 09, 2017 3:04 PM
To: wsjt-devel@lists.sourceforge.net
Cc: Ed Wilson
Subject: [wsjt-devel] FT8 Issue

 

Colleagues,

 

I have been testing FT8 since it was made available near the end of June. I
have made approximately 60 contacts with this mode and like it a lot.
JT65/JT9 seems like slow motion now! As predicted from the simulations, it
does appear to be somewhat less sensitive than JT65/JT9, but I am decoding
signals in the -19/-20 dB range with no problems.

 

I noticed an issue yesterday and I am not sure whether it might be a bug or
something at my end. I am no "spring chicken" but I think that my reflexes
are still pretty good.  Here is the issue:

 

If I call CQ and get a response as evidenced by my call on a  line with a
red background in the Band Activity window, I try to answer by
double-clicking the line. When I do this, nothing happens and I transmit
another CQ. Likewise, if I am sending CQ and I see some other station
calling CQ that I would like to contact if I double-click on the line
containing his CQ, nothing happens and I send out another CQ myself. If I
quickly disable the transmit button before double-clicking, I seem to have
no problem responding to the other station as the line moves to the RX
Frequency window and Auto Sequence can proceed..

 

I guess it is possible that I am not "quick enough on the trigger", but as I
mentioned above, I think that my reflexes are still pretty good.
Additionally, I just noticed this issue after the release of the revised FT8
protocol version on Saturday. I suppose that another possibility might be
that I need to upgrade my amateur radio computer as it is at least five
years old but it does have a quad-core Intel processor and has served me
well in all other amateur radio applications to date. Is it possible that
the decode is too slow on my computer to allow me to double-click in time? I
do not think that this is the case since I seem to have no problems if I can
quickly disable transmit before double-clicking.

 

I was not able to do a lot of testing yesterday since I had a long
"honey-do" list from my XYL, but I should have some more time today. I did
want to mention this issue before the release candidate is made available to
the "masses".

 

Thanks for any insight!

 

Ed, K0KC

 

k...@arrl.net

http://k0kc.us/

--
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 when building WSJT-X with Windows7

[Adrian Fabry]

Hi Mark,

 

Delete C:\JTSDK\src\wsjtx folder and start again build-wsjtx rinstall

 

73 Ady YO2NAA

 

  _  

From: Mark Killmon - K4SO [mailto:m...@k4so.com] 
Sent: Sunday, July 09, 2017 3:58 AM
To: wsjt-devel@lists.sourceforge.net
Subject: [wsjt-devel] Error when building WSJT-X with Windows7

 

After two clean installs of the SDK, I'm still stuck with the following
error, displayed in JTSDK-QT after an attempted "build-wsjtx rinstall."

No other errors during installation of SDK were observed.


SVN Check
---
Checking Out New Version ( wsjtx ) from SVN
---
Sourceforge Checkout Error
---
build-wsjtx was unable to checkout the
branch form Sourceforge. The service
may be down or undergoing maintenance.
Check the following link for current site
status reports:

http://sourceforge.net/blog/category/sitestatus/

Other types of errors such as non-existan branchs
or tags may alos be the casue.
Verify your entry and try again later. If the
peoblem presists, contact the wsjt-devel list.
ECHO is off.

 C:\JTSDK>


Typos are as displayed.

Can someone guide me in the right direction please?
-- 
73, Mark, K4SO
www.k4so.com 

 

--
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] Preparation for WSJT-X v1.8.0-rc1

[Adrian Fabry]

Hi Joe,

thank you and all the development team for the work.

I'd suggest two features:

1. QRA64:  Different color (or using bold chars) for the AP part of the
decoded message, when AP was used in the decode.
2. For the saved wav files, mode and submode should be included in the file
name

73 Ady YO2NAA


On Thu, Jul 6, 2017 at 6:46 PM, Joe Taylor  wrote:

> Hi all,
>
> I'm compiling a list of important new features and other changes that will
> be in WSJT-X v1.8.
>
> 1.  New mode called FT8: QSOs 4 times faster than JT9, JT65
> 2.  New mode for Frequency Calibration
> 3.  Improved performance of decoders for JT65, QRA64, and MSK144
> 4.  New utility programs: msk144code, fmeasure, fmtave, fcal, ft8sim,
> ft8d.
> 5.  Options to minimize sizes of Main window and Wide Graph window.
> 6.  New default table of recommended operating frequencies
> 7.  Improved CAT control for many rigs, including operation through
> Commander and OmniRig
> 8.  Many minor bug fixes and tweaks to the user interface
>
> Have I omitted anything that should be in this list?
>
> -- 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

[wsjt-devel] FT8 *.wav files available

[Adrian Fabry]

Hi All,

 

FT8 working nice here, a big thanks to Joe and Steve.

I have collected about 150 FT8 wav files from 20m and 6m (tropo, MS, ES).
Some of them no decode, maybe useful to analyze.

S/N from -19 to +2dB.

 

https://www.dropbox.com/sh/busdna76pjavht7/AAAXLDcC1sOfarmeLWsMU5vBa?dl=0

 

 

73 Ady YO2NAA

 

  

 

--
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] Need r7752 build for Windows XP /32

[Adrian Fabry]

Dear friends,

I would like very much to experiment with the new mode FT8, especially on 6m
multi hop. I tried to install JTSDK but it won't work on my Windows XP.

Can anyone please build an r7752 (or latest) for Windows XP SP3 / 32 bits?

Many thanks,

73 Ady YO2NAA



--
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] A couple of suggestions for consideration

[Adrian Fabry]

Hi all,

I totally agree with Bill ND0B. Here in EU  50276 is crazy, with a lot of
EU-EU QSOs during multihop openings, 99% JT65A, very few JT9 slow.
A JT65 DX window will help, imho.

On 50280 to 50292 more and more JT9 fast activity, a very good thing, so
moving MSK144 to 50260 will help.
Regarding the multi-decode for JT9 fast, I also agree, though I have a JT9E
fast sample where 3 different decodes occured, 2 from the same call +
another one (@950Hz and @575 Hz).
An option to have the vertical waterfall for fast modes would help
spreading inside the band.

Thank you for reading,

73 Ady YO2NAA



On Wed, Jun 7, 2017 at 3:33 PM, Bill ND0B  wrote:

> Good morning,
>
>
>
> With the onset of 6m Es season a couple of things have happened that lead
> to a couple of suggestions:
>
>
>
>1. The existing MSK144 calling frequency of 50.280 was found to be in
>too close proximity to the JT modes.   It was discussed and decided that,
>with the near total replacement of FSK441 it made sense to move to
>50.260.   That move is pretty much complete with only an occasional call /
>contact occurring on the old frequency.   Would it be possible to change
>the MSK144 default frequency that WSJTx ships with to 50.260?
>2. The JT mode are REALLY taking off this summer with unprecedented
>levels of activity on JT65.   This is fantastic and is leading to some
>incredible contacts however it is also somewhat of a disappointment as JT9
>fast modes are likely superior for this effort.   Monitoring ON4KST chat
>room yesterday there was a discussion of this and one of the things brought
>up that is stifling the adoption of JT9 fast modes is the lack of
>multi-decode for them.   That is a very good point and considering some of
>the comments about JT65 activity (some folks have reported over 20 decodes
>at once) this is a limitation that will keep JT9 fast modes from being
>adopted.Would it be possible to enable muti-decodes for the JT9 fast
>mode?
>
>
>
> Thank you in advance for any consideration these thought might garner.
>
>
>
> 73 de Bill ND0B
>
>
>
> 
> --
> 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