Re: [wsjt-devel] Split audio frequency

[Michael Black]

I'm sure it's not just me.  I know some tuners like SG's can take 20+
seconds to autotune a frequency.
So when responding to a CQ waiting for the tuner to do it's thing can make
your transmit message loss more likely.  My LG is typically 5 seconds or
less.  I was noticing the system retuning frequently on new QSOs and this
solves it.

And yes...I am trying to avoid the 500hz swing that occurs frequently from
QSO to QSO or, in a rare circumstance within a QSO (like somebody
responding 1499 when you CQ on 1500).  10Hz swings are much better and
don't require retuning.  If ignoring the last digits is what ALL rigs do I
guess I don't see the harm in 1Hz steps if somebody wants to pick
thatbut 10Hz is satisfactory.

I'm only doing digital right now (hearing problems) so I set up my upper
bandwidth limit to 2kHz + 200Hz to fit the JT65 signal.  Was a bit
surprised to see the rolloff occurring...thought it should've been pretty
flat up to 2200 but apparently not so I bumped it up to 2400 to fit the
1999Hz audio that can occur now.

RRR
Mike W9MDB


On Sun, Jan 3, 2016 at 1:58 PM, Bill Somerville 
wrote:

> On 03/01/2016 19:28, Michael Black wrote:
> > At least for my autotuner and rig (Omni TenTec VII and SG-237 the 10Hz
> > works quite well...the 100Hz still causes some retuning on a few bands.
>
> Hi Mike,
>
> I am still struggling with this. If you shift your DF by 1Hz from 1500
> to 1499 then the Tx VFO is moved down by 500Hz but if you move your Tx
> DF by 500Hz from 1500 to 1000 the Tx VFO still has to move 500Hz no
> matter what "split resolution" you use. Are you saying that you want to
> do smaller Tx VFO adjustments to avoid 500Hz steps when you only move
> the Tx DF by small amounts?
>
> I can see the potential gain of limiting the upper audio frequencies if
> your rig rolls off as low as 2200Hz but that seems a problem with the
> rig. SSB phone would not be very good with such a restricted bandwidth.
> Normally the Tx filter is reasonably flat between ~100Hz to 2800Hz i.e.
> a 2700Hz wide filter. If yours is rolling off as low as 2200Hz I would
> suggest the transmit carrier point is incorrectly aligned. I would check
> your Tx signal for suppressed carrier leakage!
>
> On the rig resolution situation, AFAIK they all truncate, in fact it is
> the CAT sender that usually truncates because the CAT frequency commands
> lack the lower digits. There are exceptions like the Elecraft K2 which
> ignores the last digit I believe.
>
> 73
> Bill
> G4WJS.
>
>
> --
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
--
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Spanish translation

[ea2ekh]

> On 03 Jan 2016, at 00:32, Bill Somerville  wrote:
> 
> 
> Because we are approaching a candidate release for v1.7.0, now is 
> probably not he best time to start a translation since there will be 
> many changes to the User Guide documenting the latest features. Joe K1JT 
> looks after edits of the User Guide as his name is at the top he 
> naturally wishes the content to be as he would like it. Potential manual 
> translators should coordinate with Joe on this list as to the timing of 
> a phase of translation so unnecessary duplication and reworking can be 
> minimized.

Alright :)

Awaiting instructions from our weak (signal) leader then :)


Borja.


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

Re: [wsjt-devel] Split audio frequency

[Joe Taylor]

Hi Mike,

One of the advantages of open source software is that anyone with 
special needs or desires can study the code and learn enough to make the 
changes needed to implement them.  You have done the hard work and 
evidently have solutions that suit your purposes (which, if I've 
understood correctly, include automatic fine-tuning of your 
transceiver's dial frequency and emitting male or female voice 
announcements of certain program status conditions).

Of course you should feel free to hack your copy of the code to your 
heart's content, for such purposes.  However, I see no evidence of 
widespread need or desire for such features in public versions of WSJT-X.

-- Joe, K1JT

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

[wsjt-devel] WSJT-X feature for WSPR

[Alan VK2ZIW]

Hi all,

Is there a way, for WSPR, to not submit spots for local stations?
A text file with a simple list of "ignore".

Or, am I missing something?

Alan VK2ZIW


On Sun, 3 Jan 2016 19:24:10 +, Bill Somerville wrote
> On 03/01/2016 18:58, Alessandro Gorobey wrote:
> > Now the question:
> > The decode menu (fast normal ...) have no effect on JT65 and the
> > relative parameters are now in setting -->advanced.
> > It remain valid for mode NOT jt65
> > Is this correct ?
> 
> Hi Sandro,
> 
> not quite. The menu setting "Decode->Include Correlation" is active 
> for JT65 single mode operating, it switches on the deep search 
> experienced based decoding feature that uses CALL3.TXT as a source 
> of likely to active active EME callsigns. Other decoding depth 
> settings currently have no effect with JT65 decoding.
> 
> 73
> Bill
> G4WJS.
> 
> --
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel


Alan

Man's greatest waste of time: Worshipping the wrong God.
Consider Jesus.
---
Alan Beard   Unix Support Technician from 1984 to today
70 Wedmore Rd.   Sun Solaris, AIX, HP/UX, Linux, SCO, MIPS
Emu Heights N.S.W. 2750  Routers, terminal servers, printers, terminals etc..
+61 2 47353013 (h)   Support Programming, shell scripting, "C", assembler
0414 353013 (mobile) After uni, electronics tech


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

Re: [wsjt-devel] JT65 decoder parameters

[Alessandro Gorobey]

Hi Bill,

Il 03/01/2016 20:24, Bill Somerville ha scritto:
> On 03/01/2016 18:58, Alessandro Gorobey wrote:
>> Now the question:
>> The decode menu (fast normal ...) have no effect on JT65 and the
>> relative parameters are now in setting -->advanced.
>> It remain valid for mode NOT jt65
>> Is this correct ?
>
> Hi Sandro,
>
> not quite. The menu setting "Decode->Include Correlation" is active for
> JT65 single mode operating, it switches on the deep search experienced
> based decoding feature that uses CALL3.TXT as a source of likely to
> active active EME callsigns. Other decoding depth settings currently
> have no effect with JT65 decoding.

Thank, now understand.
I confuse the 5 values of dec_data.params.ndepth with the 5 bits of
dec_data.params.nexp_decode

73
Sandro
IW3RAB



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

Re: [wsjt-devel] Split audio frequency

[Black Michael]

I'll agree that the voice stuff is a "nice to have".  But maybe you can explain 
the logic of the 500Hz window100Hz would seem to make more sense and I 
don't quite see the logic of 500Hz when it causes tuners to have to retune.  
when changing offsets.   Why 500 vs 100?  That's a one-line change.  100Hz much 
less retuning (probably zero)...500Hz frequent retuning...

I'm not going to keep changes in my code that mess with the ui code since it 
really messes things up with merges so I'll just revert to non-gui changes.

RRRMike W9MDB


  From: Joe Taylor 
 To: WSJT software development  
 Sent: Sunday, January 3, 2016 7:41 PM
 Subject: Re: [wsjt-devel] Split audio frequency
   
Hi Mike,

One of the advantages of open source software is that anyone with 
special needs or desires can study the code and learn enough to make the 
changes needed to implement them.  You have done the hard work and 
evidently have solutions that suit your purposes (which, if I've 
understood correctly, include automatic fine-tuning of your 
transceiver's dial frequency and emitting male or female voice 
announcements of certain program status conditions).

Of course you should feel free to hack your copy of the code to your 
heart's content, for such purposes.  However, I see no evidence of 
widespread need or desire for such features in public versions of WSJT-X.

    -- Joe, K1JT

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


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

Re: [wsjt-devel] Fwd: Kvasd installer

[Greg Beam]

Hello,

I was under the impression, and maybe this is wrong, Fedora uses Sudo by 
default.

In any case, unless a user is well versed in using root on a system 
implemented with Sudo, they should not be using root as it will cause 
all sorts of problems. I have no desire to mix root and sudo commands in 
this script as it will no doubt lead to problems for users on a sudo 
system installation. Personally, I have no problem using either 
implementation, but those new to Linux or using these packages for the 
first time could find themselves in a mess rather quickly.

I'm not going to debate the merits of using or not using Root / Sudo, as 
that is up to the user to determine, however, with the next release of 
WSJT-X (v1.7.x series), it is expected KVASD will not longer be needed, 
as such, kvasd-installer will no longer be needed. At the time of it's 
writing, the new FOSS decoder was not yet on the drawing board, and 
kvasd-installer was a work around for packaging issues.

If you want to use Root v.s. Sudo, simply search and replace the calls 
or remove the sudo prefix commands completely and call the script from 
the root user directly.


73's
Greg, KI7MT

On 1/1/2016 04:26, Bill Somerville wrote:
> Hi All,
>
> forwarding this here as I'm not sure if Greg or Richard are best able to
> answer Saku's questions.
>
> 73 & HNY
> Bill
> G4WJS.
>
>
>  Forwarded Message 
> Subject:  Kvasd installer
> Date: Fri, 01 Jan 2016 11:09:40 +0200
> From: Sakari Nylund 
> Reply-To: sakari.nyl...@nic.fi
> To:   Bill Somerville 
>
>
>
> Hi Bill!
>
> Happy new year!
>
> This might not be your target, but as a member of development team I'll
> inform you about kvasd-installer.
>
> I just installed a fresh copy of Fedora 23 (32bit) to old laptop and now
> I'm filling it up with software I'm going to use.
>
> Noticed that kvasd-installed does not work as said in wsjt-x 1.6.0 manual.
>
> There are 2 snapshots included.
>
> Problem is that not every user is included into sudoers list. I
> personally do not use sudo at all. I always do su (+root password)
> when needed, as cracking user password (who is in sudoers list) may lead
> also to root access.
>
> It took few seconds for me to find where in heavens kvasd binary is as
> download (after first error shown) went ok .
> Regular user may have more difficulties with this.
>
> I suggest a line "Give username for user in sudoers list, or use root"
> and next line "give password"
> rather than "[sudo] password for person "
>
> As it is impossible question if user  is not included to sudoers list.
>
>
> --
> Saku
> OH1KH

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

Re: [wsjt-devel] Split audio frequency

[Bill Somerville]

On 03/01/2016 14:40, Michael Black wrote:
> The attached patch adds a "Freq Resolution" combo box for split 
> resolution.  It defaults to the current value.
> When using an auto tuner changing offset in split mode can cause the 
> tuner to retune since the audio can start anywhere in the 1500-2000Hz 
> range.  On my rig I can use 1Hz and the tuner stays  happy on any 
> offset and retuning on band changes is instantaneous now since it 
> doesn't need so many memories for all the different base audio 
> frequencies possible with 500hz.
>
> RRR
> Mike W9MDB
Hi Mike,

I don't understand the purpose of this change. The Tx dial frequencies 
are constrained by the requirement to deliver 1500-2000Hz audio to the 
transmitter modulator, how does this change make a difference?

Aside from that I think you need to limit the smallest resolution to 
100Hz since some rigs only tune in coarse steps. There are several rigs 
with 10Hz steps and I believe there are one or two with 100Hz minimum steps.

73
Bill
G4WJS.

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

Re: [wsjt-devel] Spanish translation

[Anders Rhodin]

Hi Bill and the rest,

QSL on the changes in the script and user guide location. 
I´ve made my first steps in learning asciidoc yesterday and also made some test 
building of the user guide using a short translated part just to try it out. No 
problems yet!  

As you said, it´s maybe not the best time to make a rush for a translation for 
the v1.7.0 candidate release. My suggestion is to wait until the English user 
guide for the candidate release is ready and start of from there. 
 
However, I´ll leave the final decision to Joe to decide when and on what user 
guide to start translating from.
Meanwhile, I´ll try to get in shape with my translating skills and practice 
with asciidoc.  

73 de SM7VRZ, Anders


Från: Bill Somerville
Skickat: den 3 januari 2016 05:11
Till: wsjt-devel@lists.sourceforge.net
Ämne: Re: [wsjt-devel] Spanish translation

On 02/01/2016 14:10, Bill Somerville wrote:
> The easiest way to work with the user guide is to install the JTSDK so
> you can build WSJT-X. That will allow you to build the User Guide. I
> will make some changes to the build script to allow for multiple
> language User Guides so you will have a simple single command to
> generate the document to check out your changes.
Hi Anders & Borja,

I have made some changes to WSJT-X and its build script to cater for 
multiple language User Guides. Currently only one User Guide is built 
which is the English version. I only need to make a one line change to 
add a new language when we have some translated sources. It relies on 
the manual source files having exactly the same names for each language. 
Source files will be in the following places:

   -->   doc  -->  user_guide  --> en  - wsjtx-main.adoc
...
images  -->  image PNG files

A translated version of the sources will reside in a sibling of the 'en' 
directory named with the lower case two letter ISO 639 language code 
e.g. 'es' for Spanish, 'sv' for Swedish. Optionally country level 
localization is catered for by appending the upper case two or three 
letter ISO 3166 country code e.g. en_GB but I doubt they will be needed.

Because we are approaching a candidate release for v1.7.0, now is 
probably not he best time to start a translation since there will be 
many changes to the User Guide documenting the latest features. Joe K1JT 
looks after edits of the User Guide as his name is at the top he 
naturally wishes the content to be as he would like it. Potential manual 
translators should coordinate with Joe on this list as to the timing of 
a phase of translation so unnecessary duplication and reworking can be 
minimized.

73
Bill
G4WJS.

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

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

[wsjt-devel] Split audio frequency

[Michael Black]

The attached patch adds a "Freq Resolution" combo box for split
resolution.  It defaults to the current value.
When using an auto tuner changing offset in split mode can cause the tuner
to retune since the audio can start anywhere in the 1500-2000Hz range.  On
my rig I can use 1Hz and the tuner stays  happy on any offset and retuning
on band changes is instantaneous now since it doesn't need so many memories
for all the different base audio frequencies possible with 500hz.

RRR
Mike W9MDB


splitres.patch
Description: Binary data
--
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] JT65 decoder parameters

[Alessandro Gorobey]

Hi All,

I working with r6330 decoder.

It work absolutely fine, decode also signals in high noise environment.
Seems also is deaf to problems caused by nasty transmissions.
Also using for experiment max values for aggressive decoding and random 
erasure never got program error and false decoding are very limited.

Now the question:
The decode menu (fast normal ...) have no effect on JT65 and the 
relative parameters are now in setting -->advanced.
It remain valid for mode NOT jt65
Is this correct ?

Compliments to Joe and Steve for the superlative decoder

73 && HNY
Sandro
IW3RAB

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

Re: [wsjt-devel] WSJT-X r6333/r6334 WSPR-2 Hang

[Claude Frantz]

On 01/02/2016 01:38 PM, Josh Rovero wrote:

> Both built from source, Linux x86_64.  WSPR-2 stops
> processing (hangs) with steady decode indicator after
> a couple of hours of operation.  Main window and
> spectrum GUIs remain open, but unresponsive.
>
> I start wsjtx from a console window, no obvious error
> indications there.

Hi you all,

At frist: Happy New Year !

I have observed a similar behavior on some occasions on a 32 bit Linux 
system, on other modes. After some examination, I suspected a problem 
related to ALSA. An interesting additional observation was that the 
problem did not occur when pavucontrol was running. Although the sound 
system is pulseaudio, Qt5 seems to come via an ALSA Interface.

Please excuse me for this imprecise diagnostic.

Best 88 de Claude (DJ0OT)

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

Re: [wsjt-devel] Split audio frequency

[Michael Black]

At least for my autotuner and rig (Omni TenTec VII and SG-237 the 10Hz
works quite well...the 100Hz still causes some retuning on a few bands.
But why not 1Hz too?  Is that going to harm something?  Seems to me if the
step size doesn't work for your rig...don't use it.  Lowest common
denominator should only apply where no options are available.
And since tuning is done at 1500Hz seems the target split offset should be
the same if possible.
The 500 window is too wide to keep my tuner happy.  I also noticed that
since I had my bandwidth set at 2200Hz that the upper end was apparently
rolling off power somewhat.
What happens if you pass a more-than-acceptable frequency to one of those
rigs?  Does it reject it or round it off?  I'd think it would just round it
or truncate it.

RRR
Mike W9MDB


On Sun, Jan 3, 2016 at 9:05 AM, Bill Somerville 
wrote:

> On 03/01/2016 14:40, Michael Black wrote:
> > The attached patch adds a "Freq Resolution" combo box for split
> > resolution.  It defaults to the current value.
> > When using an auto tuner changing offset in split mode can cause the
> > tuner to retune since the audio can start anywhere in the 1500-2000Hz
> > range.  On my rig I can use 1Hz and the tuner stays  happy on any
> > offset and retuning on band changes is instantaneous now since it
> > doesn't need so many memories for all the different base audio
> > frequencies possible with 500hz.
> >
> > RRR
> > Mike W9MDB
> Hi Mike,
>
> I don't understand the purpose of this change. The Tx dial frequencies
> are constrained by the requirement to deliver 1500-2000Hz audio to the
> transmitter modulator, how does this change make a difference?
>
> Aside from that I think you need to limit the smallest resolution to
> 100Hz since some rigs only tune in coarse steps. There are several rigs
> with 10Hz steps and I believe there are one or two with 100Hz minimum
> steps.
>
> 73
> Bill
> G4WJS.
>
>
> --
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
--
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] OS X 10.11.3 success

[George J Molnar]

For Bill and the Mac side of the house…

Used the new build scripts under OS X 10.11.3 with smooth and error-free 
results for 6344.

Only changed OSX_DEPLOYMENT_TARGET and OSX_SYSROOT to match my system. 

Thanks, Bill!

George J Molnar
KF2T | AFA9GM
Twitter: @GJMolnar

SUPPORT HR-1301 & S-1685
http://www.arrl.org/amateur-radio-parity-act


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

Re: [wsjt-devel] Split audio frequency

[Bill Somerville]

On 03/01/2016 19:28, Michael Black wrote:
> At least for my autotuner and rig (Omni TenTec VII and SG-237 the 10Hz 
> works quite well...the 100Hz still causes some retuning on a few bands.

Hi Mike,

I am still struggling with this. If you shift your DF by 1Hz from 1500 
to 1499 then the Tx VFO is moved down by 500Hz but if you move your Tx 
DF by 500Hz from 1500 to 1000 the Tx VFO still has to move 500Hz no 
matter what "split resolution" you use. Are you saying that you want to 
do smaller Tx VFO adjustments to avoid 500Hz steps when you only move 
the Tx DF by small amounts?

I can see the potential gain of limiting the upper audio frequencies if 
your rig rolls off as low as 2200Hz but that seems a problem with the 
rig. SSB phone would not be very good with such a restricted bandwidth. 
Normally the Tx filter is reasonably flat between ~100Hz to 2800Hz i.e. 
a 2700Hz wide filter. If yours is rolling off as low as 2200Hz I would 
suggest the transmit carrier point is incorrectly aligned. I would check 
your Tx signal for suppressed carrier leakage!

On the rig resolution situation, AFAIK they all truncate, in fact it is 
the CAT sender that usually truncates because the CAT frequency commands 
lack the lower digits. There are exceptions like the Elecraft K2 which 
ignores the last digit I believe.

73
Bill
G4WJS.

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

Re: [wsjt-devel] OS X 10.11.3 success

[Bill Somerville]

On 03/01/2016 19:54, George J Molnar wrote:
> Used the new build scripts under OS X 10.11.3 with smooth and 
> error-free results for 6344.
>
> Only changed OSX_DEPLOYMENT_TARGET and OSX_SYSROOT to match my system.

Hi George,

Well done! Sorry about the slightly outdated instructions for installing 
CMake on OS X, I have updated them.

You do not really need to change the OSX_DEPLOYMENT_TARGET as it is a 
minimum value but it does no harm.

73
Bill
G4WJS.

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