Re: [wsjt-devel] choosing the best CRC for short data blocks

2018-01-26 Thread Gary McDuffie 


> On Jan 25, 2018, at 10:26 PM, Paul Black via wsjt-devel 
>  wrote:
> 
> And when I can hear it, see it on the waterfall display and it doesn't decode

Frequently caused by computer clock error, but I assume you probably already 
have that under control.

Gary - AG0N
--
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] tx audio equalisation, just a thought.

2018-01-26 Thread Libor Holouš 
Sorry Mike.. This was directed to Peter of course.. :-)
  - Original Message - 
  From: Libor Holouš 
  To: WSJT software development 
  Sent: Friday, January 26, 2018 12:31 PM
  Subject: Re: [wsjt-devel] tx audio equalisation, just a thought.


  Hello Mike, --
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] tx audio equalisation, just a thought.

2018-01-26 Thread Libor Holouš 
Hello Mike, 
just for sure, have you checked, that your sound card doesn't use equalising 
settings?
Some soundcards do have this possibility and in my PC serviceman career I have 
seen many times, that people complained for audio quality in Skype and finaly 
they had switched some DSP effects, like "bathroom" on. Typicaly it was Realtek 
HD audio.. But some other have this functionality too. (If not using USB 
dongle. :-)
If not and you have this possibility, it could solve your issue a bit, because 
you could create some own profile and use equaliser and W-mtr to calibrate all 
subaudioband to one power level (using tune button and clicking by watterfall). 
Of course not as great as you suggest, but can help a bit.

See my IDT audio panel:



73 Libor
  
- Original Message - 
From: "Bill Somerville" 
To: 
Sent: Friday, January 26, 2018 11:49 AM
Subject: Re: [wsjt-devel] tx audio equalisation, just a thought.


> On 26/01/2018 10:28, Peter Sumner wrote:
>> Hello Mike,
>> can you offer some URL's as I have exhausted my google searches for 
>> both of those suggestions... all I get is topic about fishing for rig 
>> split and fake it gives me pages of topics about the US president and 
>> his fake news. have tried various other combinations but as I do not 
>> know what they are meant to do it is hard to refine the searches.
>>
>> Peter.
> 
> Hi Peter,
> 
> I think Mike may have missed that you are using EME on the microwave 
> bands. The facility he suggests is only appropriate and available for 
> the narrower modes used commonly on HF.
> 
> With respect to your original request, I do see some merits but 
> profiling your Tx pass band characteristics for a mode like QRA64 
> although it is quite tricky given the number of frequencies that need 
> calibration. The implementation would not be too hard as a simple table 
> of factors would be mapped to each tone frequency within the digital 
> audio synthesizer, OTOH the work to implement a maintenance facility for 
> the correction factors would be considerable and I doubt considered 
> worth the payback. Also modern rigs are dealing with this more 
> fundamentally by providing SSB-DATA modes tailored for AFSK operation 
> which have very flat pass band characteristics.
> 
> 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] tx audio equalisation, just a thought.

2018-01-26 Thread Bill Somerville 

On 26/01/2018 10:28, Peter Sumner wrote:

Hello Mike,
 can you offer some URL's as I have exhausted my google searches for 
both of those suggestions... all I get is topic about fishing for rig 
split and fake it gives me pages of topics about the US president and 
his fake news.  have tried various other combinations but as I do not 
know what they are meant to do it is hard to refine the searches.


Peter.


Hi Peter,

I think Mike may have missed that you are using EME on the microwave 
bands. The facility he suggests is only appropriate and available for 
the narrower modes used commonly on HF.


With respect to your original request, I do see some merits but 
profiling your Tx pass band characteristics for a mode like QRA64 
although it is quite tricky given the number of frequencies that need 
calibration. The implementation would not be too hard as a simple table 
of factors would be mapped to each tone frequency within the digital 
audio synthesizer, OTOH the work to implement a maintenance facility for 
the correction factors would be considerable and I doubt considered 
worth the payback. Also modern rigs are dealing with this more 
fundamentally by providing SSB-DATA modes tailored for AFSK operation 
which have very flat pass band characteristics.


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] question and/or feature request

2018-01-26 Thread Georg Isenbürger 
David, 

if you want to see what is going on during time sync, you can use 
„terminal.app" in your Mac:

open it and type:

sudo ntpdate -u time.apple.com


Hit ENTER

you will be asked for a password which is the password you use to login. You 
need to be an administrator.

It takes a couple of seconds. The time on your mac will be adjusted and it also 
shows how much is was corrected.


73s
Georg, DJ6GI, NZ1C


Am 26.01.2018 um 11:19 schrieb Bill Somerville :

HI David,

the NTP time sync protocol doesn't just keep your machine synchronized with an 
accurate Internet time source, it also tries to guarantee monotonic time 
increments. This can mean that when the inaccuracy is a lot, perhaps after an 
extended period with no network connection or shut down, the clock cannot be 
stepped directly to the server time. For example the monotonicity requires that 
backward steps are not allowed so tiny adjustments must be made between 
internal clock ticks until synchronization is achieved. When you go to the Date 
and Time preferences panel macOS assumes you are happy to accept a large step, 
backwards or forwards, and will correct the time immediately.

73
Bill
G4WJS.

On 26/01/2018 02:51, David Jaffe wrote:
> That must have been it. That’s how superstitions are born! ( Odd though, 
> because I had network time synch enabled.  )
> 
> On Jan 25, 2018, at 8:24 AM, Jim Record  > wrote:
> 
>> When you changed the time zone, OSX updated your clock. This probably got 
>> you working. No reason to use anything but your local time zone.
>> 
>> Jim
>> AD0YO 
>> 

--
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] choosing the best CRC for short data blocks

2018-01-26 Thread Bill Somerville 

On 26/01/2018 05:26, Paul Black via wsjt-devel wrote:
And when I can hear it, see it on the waterfall display and it doesn't 
decode, It would be good to have the option of knowing why. I would 
like a display showing the data received from the packet and an 
indication of how far it got through the error correction, 
decompression and decoding process.


I wonder if anyone else would be interested in this level of detail?


Hi Paul,

that might be more difficult than you think. WSJT-X does give a bit more 
information for some modes, for example the modes commonly used for EME 
have a mechanism to display the detection of sync patterns, the first 
step in decoding a signal, and the "quality" of that sync detection. The 
MSK144 decoder provides some statistics of the number of frames 
averaged, the number of bits corrected by FEC, and the coherence of the 
signal being decoded, but this information is only provided for 
successful decodes.


In modes like WSPR and those used commonly on HF for QSOs, where 
multiple signals are being decoded, there would be far too much 
information to be useful, if for example we listed every decode attempt 
that yielded no decode there would be hundreds or even thousands of 
events to report per good decode. In general it is probably only 
practical to provide a bit of quality information with the successful 
decodes. The steps you list above are always completed if a decode 
attempt passes the decoding threshold apart from the critical first 
error correcting reconstruction step. Before that step the data is just 
a big number with little relationship to a human readable, or useful, 
message.


Related to this is one of the decoder algorithms used in WSJT-X, 
particularly for FT8 and MSK144, that always yields a decoded message 
which is then filtered for correctness by checking a checksum embedded 
in the message. So the obvious statistic of checksum failed would occur 
for every bit of noise, QRM , or birdy that invoked a decoding attempt.


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] tx audio equalisation, just a thought.

2018-01-26 Thread Peter Sumner 
Hello Mike,
 can you offer some URL's as I have exhausted my google searches for both
of those suggestions... all I get is topic about fishing for rig split and
fake it gives me pages of topics about the US president and his fake news.
have tried various other combinations but as I do not know what they are
meant to do it is hard to refine the searches.

Peter.

On Fri, Jan 26, 2018 at 3:30 PM, Black Michael via wsjt-devel <
wsjt-devel@lists.sourceforge.net> wrote:

> You run either rig split or fakeit.  That keeps all your audio in the
> 1500-2000 range.
>
> de Mike W9MDB
>
>
>
> On Thursday, January 25, 2018, 7:02:45 PM CST, Peter Sumner <
> vk8...@gmail.com> wrote:
>
>
> Hello,
>  first off, sorry if this subject has already been discussed and resolved
> but could not see anything on my searches.
>
> As the owner of an older Icom VHF SSB rig, I have a seemingly common
> problem that my TX pass band is not flat, in fact using QRA64-D earlier
> today with Rex, VK7MO revealed a significant ripple across the tones (about
> 3db) seen on the Bird output power meter.
> As I am unsure how much control there is over the level of the tones
> generated by WSJT-X but I wondered if there is a way to build in a simple
> level control for TX tones across the passband?
> In my head I had thought an adjuster for 500-750 750-1000 1000-1250 and so
> on and be able to change the nominal value from 0db (flat) to be -1db or
> +1db for each segment of the decided passband...
>
> Maybe others have a better way of managing this that is not part of WSJTX?
>
> Regards,
> Peter, vk5pj
> 
> --
> 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] question and/or feature request

2018-01-26 Thread Bill Somerville 

HI David,

the NTP time sync protocol doesn't just keep your machine synchronized 
with an accurate Internet time source, it also tries to guarantee 
monotonic time increments. This can mean that when the inaccuracy is a 
lot, perhaps after an extended period with no network connection or shut 
down, the clock cannot be stepped directly to the server time. For 
example the monotonicity requires that backward steps are not allowed so 
tiny adjustments must be made between internal clock ticks until 
synchronization is achieved. When you go to the Date and Time 
preferences panel macOS assumes you are happy to accept a large step, 
backwards or forwards, and will correct the time immediately.


73
Bill
G4WJS.

On 26/01/2018 02:51, David Jaffe wrote:
That must have been it. That’s how superstitions are born! ( Odd 
though, because I had network time synch enabled.  )


On Jan 25, 2018, at 8:24 AM, Jim Record > wrote:


When you changed the time zone, OSX updated your clock. This probably 
got you working. No reason to use anything but your local time zone.


Jim
AD0YO



--
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