On Jul 7, 2015, at 7:23 AM, f5djl <[email protected]
<mailto:[email protected]>> wrote:
Hi Steve , Hi Joe and all
A few more details following some furthers tests today :
1/ T0 : start application enable TX ticked (red) ( nothing else touch
ie no modification of Tx% at 20% )
2/ T0+1h : Happily decoding but no TX
3/T0+2h : same status
4/T0+2h10mn : press TX next , one TX cycle and then no further TX
5/T0+3H : still no further TX
6/T0+3H20 : Change TX & at 25 %
7/ As at still no Tx the press TX_next (09h48) and then
process started :-)
8/ Now more than 2 hours later still running happily for how
long … but so far so good ! ?
You will find screenshots and log files at dropbox below .
https://www.dropbox.com/s/w2f013omrzc9lu8/WSPR%20test%20scheduler.zip?dl=0
My conclusion so far is that to ensure start of the scheduler : Press
enable TX , followed by a change of percentage AND then a TX next
is required !
Once started will it stop? I do not know (yet) but it has been
running more than two hours now ! :-)
I may try to have look at code :WsprTxScheduler, WSPR BandHopping and
mainwindows I assume , but I am sure you will be more efficient to
correlate it with above . ( I am much more a C microcontroller man
or financial market more than a QT expert :-) )
Very best regards
Jean Louis
Test environment : Windows 8.1 , WSJT-X r 5685 , FT897
_Summary to Steve's questions : _
Does this happen only if you have never touched TX Pct since starting
the program?
>Yes
When you observed this behavior what was the TX Pct?
>20 %
How long did you run with no transmissions?
>More than 2 Hours
In particular, did you go for a full 2 hours with no transmission?
>Yes
If so, then does the scheduler eventually kick in after 2 hours with
no transmission?
>Only triggered by a TxNext AND a change of %
-----Message d'origine-----
De : Steven Franke [mailto:[email protected]]
Envoyé : mardi 7 juillet 2015 02:17
À : WSJT software development
Objet : Re: [wsjt-devel] WSPR-2 - Scheduler in no band hopping mode
Jean Louis, Joe -
Does this happen only if you have never touched TX Pct since starting
the program?
When you observed this behavior what was the TX Pct? How long did you
run with no transmissions? In particular, did you go for a full 2
hours with no transmission? If so, then does the scheduler eventually
kick in after 2 hours with no transmission?
I haven’t changed anything in the wspr tx scheduler recently. I know
that Bill adjusted the band selection algorithm not too long ago, but
that shouldn’t have involved the scheduler. I just looked at the
code, and it seems that the tx table should be automatically
generated the first time that next_tx_state() is called.
I’ll also note that the statistics of the scheduler may “feel”
different when TX Pct is less than 17%, as a different algorithm is
used in that regime.
Steve k9an
> On Jul 6, 2015, at 6:32 PM, Joe Taylor <[email protected]
<mailto:[email protected]>> wrote:
>
> Hi Jean Louis and all,
>
> I have not yet tried to trace the cause, but I have also observed the
> behavior you describe. In WSPR mode, WSJT-X can stop initiating any
> Tx sequences, even though "Enable Tx" remains ON (red background).
>
> I think Bill (or was it Steve?) made the most recent changes to the
> T/R scheduler. Is it possible that somehow the T/R table becomes
> lost, or is not re-calculated when that should happen?
>
> -- Joe, K1JT
>
> On 7/6/2015 9:20 AM, f5djl wrote:
>> Hello to all
>>
>> While testing the patch kindly provided by Mike for the rig control
>> of FT991 ( thanks Mike and it seems to work smoothly for WSPR/991) ,
>> I noticed the following (BTW observed on two different TX
>> configurations FT897 and FT
>> 991) :
>>
>>1/WSPR seems stuck in RX mode despite the enable TX button being red .
>>
>>2/ Just changing the TXPct % seems to resume the transmit process
>> ( I understand this action forces to recalculation of the table used
>> by the scheduler )
>>
>>3/ I tried a TX next button before changing the TX pct , and
>> although it triggered a single transmission it did not resume the
scheduler.
>>
>>4/ I have not yet managed to figure what could have stopped TX
>> process , is it the automatic ( every two hours ?) recalculation not
>> happening or something else not sure .
>>
>>5/ I have seen it working for several hours without problem so it
>> is not always the case .
>>
>> Any advice on how to better capture the condition which lead to TX
>> process stopping will be much appreciated ( I had a look at
>> WSPR_history file and except the absence of TX slot could not see
>> anything obvious)
>>
>> Latest tests were performed on Win 8.1 , 1.6.0-dev r 5685 with FT897
>> and
>> FT991 .
>>
>> Very best regards
>>
>> Jean Louis F5DJL
>>
>>
>>
>> PS : If anybody has hand-on experience with FT991 in DATA-USB mode
>> thru the TRX internal audio card ( ie audio via internal USB
>> interface ) , please contact me direct as the setting of the DSP
>> filter in that specific mode ( Data-USB not USB) on the unit at hand
>> does not perform as per the page 58 the manual ! I will be interested
>> to share experience with another owner of FT 991) . I have also
>> joined FT991 group as this is not specific to WSJT modes.
>
