Is it ok to define Task and Task Rules from PRF file instead of TSK file?

[DPLuigi]

Hello,

I am working with Simon Taylor/Staylo to update his code XCondor which generates the XCSoar configuration files form a Condor FPL flight plan file. At the moment the task parameters (sector/cylinder parameters, ...)n are not configured.

I noticed that all the Task parameters (Config/Setup System/ Menu 13) and the Task Rules (Config/Setup System/ Menu 14) are saved in the PRF file.  Similarly with initial forecast for wind speed and direction.

I am just trying to understand teh format of the TSK file (memory dump of Task parameters) from source code but it is not easy for me as I just started and I am still getting my mind around this.  So, is there a difference in setting up a Task from parameters set in a PRF file rather than using TSSK file?

Is there issues with using PRF file as a work around to set the task parameters?

Note that, as I understand it, XCondor makes use of the fact that the TP coordinates from a generic/pre-saved TSK file are updated by those from a waypoint file, if it is presents with the same turnpoint name.

Is there someone who could describe me the format of of the PRF if I wanted to write/generate my own PRF binary file  from FPL?

On Condor Club, Thierry/BOD1 did implement such tool, only available for task download from there, and this TSK file only contains Config/Settings/Menu 13 parameters and not Task rule parameters from Config/Settings/Menu 14.
See this link for this feature: http://www.condor-club.eu/showtask/0/?id=1815

Regarding the surface temperature Max temp and glider ballast, although they seemed to be saved and retrieved properly for the next XCSoar session, I could not find where there were saved.  Any suggestions?

Also, regarding the Sunset time calculated with Sun Ephemeris by XCSoar (see manual section 4.16 p.79) . For Condor online session, the task can be set at any date/season and time and thus the system time and date are irrelevant.  It is clear that only the GPS data from Condor can make the date and time consistent between simulation and XCSoar.  I noticed the the Sunset times are way off for the date I flew Condor tasks.  For example for task flown in the Czech Republic on June 21st, 2009, the Sunset is a 4.11pm!!!
Is XCSoar retrieving date from GPS when Time is obtain from GPS?  Or is it a bug?

Thank you in advance for you clarifications and suggestions.

Donat-Pierre Luigi


------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

[DPLuigi]

From a different thread Rob pointed this:

Rob Dunning wrote:

You'd asked earlier about the task files for Condor.  XCSoar's task rules and parameters are stored as defaults in the PRF file, but these same parameters are also stored in the .TSK file and the values in the .TSK file override those in the PRF file when a task file is loaded.  So, you may have to write to the binary .TSK file -- see the LoadNewTask() and SaveTask() functions to see how XCSoar stores it.  Alternatively, there was a recent request for functionality to import tasks into XCSoar from other file types.  What other task file types can your application write?

I tested again this approach using an edited PRF files as I described in my first post and it works...  The TSK files provided with with XCondor is a workaround that Simon Taylor used to define a task waypoints, i.e. they are generic.

After starting XCSoar with an edited PRF file, or from a previous session, with the desired parameters in menu 13 and 14, the selection of TSK file for flight plan with 1,2, ..., 8 waypoints between start and finish did not change these parameters.  Maybe because those TSK files provided with/for XCondor are "dummy" TSK files, they did not override those parameters in menu 13 and 14 set at start-up from a PRF file.

I guess I am double checking where with other XCSoar users on this list that the work-around I found does work, and was also observed by others.  It is a simple approach to get these parameters from menu 13 and 14 uploaded without relying on a specific TSK file.  I agree with Rob though this is a hack...  Anyhow, Simon will have a go at it with his XCondor application.  I'll start looking too at the TSK file read and write functions in XCSoar source code.

Regarding the file type used by condor to define a task it is a text file, with extension *.fpl, which was XCondor use to set-up most of what XCSoar parse to display the right map, start airport, waypoints, glider polar, ....  the only thing missing are the task parameters (Task rules, waypoint type and min/max altitudes) and the initial wind estimate  (i.e. direction and speed).

I will update this list when there are some progress with implementing this in XCondor.

Thanks,
Donat

[John Wharington-2]

Hi Donat,

Remember that XCSoar is primarily driven by the needs of real-life
pilots rather than Condor, and the constraints/drivers from Condor may
not be appropriate for XCSoar's primary function.

Condor itself is "guilty" of not having everything neatly organised; for
example the ballast setting is not saved in the .fpl file.

By the way, I have been testing the sunset algorithm and find it works
correctly if it is fed the correct GPS date and has the correct timezone
offset for the corresponding location.

By this I suggest that possibly Condor might always send out today's
date; and secondly that your use of -7 for the timezone corresponding to
Czech is wrong.

See here the output from the sunset algorithm, which I extracted from
the sourcecode and turned it into a standalone test program:

 Sunrise and set
===============
  year  : 2009
  month : 6
  day   : 20

Days since Y2K :  3458
Latitude :  50.1, longitude: 14.5, timezone: 2.0
Declination   :  23.44
Daylength     : 16:21 hours

Civil twilight: 4:13
Sunrise       : 4:53
Sun altitude  63.34 degr South at noontime 13:3
Sunset        : 21:14
Civil twilight: 21:53


On Sat, 2009-02-21 at 19:50 -0800, DPLuigi wrote:

> >From a different thread Rob pointed this:
>
>
> Rob Dunning wrote:
> >
> > You'd asked earlier about the task files for Condor.  XCSoar's task rules
> > and parameters are stored as defaults in the PRF file, but these same
> > parameters are also stored in the .TSK file and the values in the .TSK
> > file override those in the PRF file when a task file is loaded.  So, you
> > may have to write to the binary .TSK file -- see the LoadNewTask() and
> > SaveTask() functions to see how XCSoar stores it.  Alternatively, there
> > was a recent request for functionality to import tasks into XCSoar from
> > other file types.  What other task file types can your application write?
> >
>
> I tested again this approach using an edited PRF files as I described in my
> first post and it works...  The TSK files provided with with XCondor is a
> workaround that Simon Taylor used to define a task waypoints, i.e. they are
> generic.
>
> After starting XCSoar with an edited PRF file, or from a previous session,
> with the desired parameters in menu 13 and 14, the selection of TSK file for
> flight plan with 1,2, ..., 8 waypoints between start and finish did not
> change these parameters.  Maybe because those TSK files provided with/for
> XCondor are "dummy" TSK files, they did not override those parameters in
> menu 13 and 14 set at start-up from a PRF file.
>
> I guess I am double checking where with other XCSoar users on this list that
> the work-around I found does work, and was also observed by others.  It is a
> simple approach to get these parameters from menu 13 and 14 uploaded without
> relying on a specific TSK file.  I agree with Rob though this is a hack...
> Anyhow, Simon will have a go at it with his XCondor application.  I'll start
> looking too at the TSK file read and write functions in XCSoar source code.
>
> Regarding the file type used by condor to define a task it is a text file,
> with extension *.fpl, which was XCondor use to set-up most of what XCSoar
> parse to display the right map, start airport, waypoints, glider polar, ....
> the only thing missing are the task parameters (Task rules, waypoint type
> and min/max altitudes) and the initial wind estimate  (i.e. direction and
> speed).
>
> I will update this list when there are some progress with implementing this
> in XCondor.
>
> Thanks,
> Donat

------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

[DPLuigi]

Hi John,

I hope you had a nice week-end and found some time to fly this week-end.  You are in Australia, aren't you? I remember vaguely that is was inferred in a post somewhere.
 

John Wharington-2 wrote:

Remember that XCSoar is primarily driven by the needs of real-life
pilots rather than Condor, and the constraints/drivers from Condor may
not be appropriate for XCSoar's primary function.

I am aware of that and I thought I clearly expressed my concerned to not be too off topic.  I thought also I warned at the beginning of my post(s) that this was about issues when using XCSoar with Condor.  Anyhow, I hope I am now disruptive with my posts on this forum.  Please let me know.

A little about where I come from... I learned to fly for real in the UK, during my Ph.D. at UMIST in the UK and I solo-ed there about 14 years ago - overall flying for about 2 years.  
I was fortunate to meet very nice and welcoming members of the local soaring club, LVVSA, Jean (Nevada), and was able to fly as a P2 twice since Otcober. We all hope I can join as soon as my finance allows me to fly regularly, and I would love to re-solo and start where I left off 13 years ago fly X-country.

Condor is a very fun tool/flight simulator to keep my mind into it but you are right, it is a lame comparison to the real experience I miss very much.  I am very enthusiastic about finding out how XCSoar helps with the work load during XCountry and flight planning.  It is also a sophisticated analytical tool that I am learning to enjoy and understand in the process.  I hope to be seasoned with it by the time I start X-country flying, which could be within a year (or two) and once I solo-ed and flew for a while from the "seats of my pants".  In the mean time, Condor is a cheap way to fulfill a long standing passion that we all share here.

Condor itself is "guilty" of not having everything neatly organized; for
example the ballast setting is not saved in the .fpl file.

I am learning this too while looking with Simon Taylor how to help prepare XCSoar configuration from Condor flight plans but also looking into the Free Condor Instrument Suite, which suffers form not having access to current weight (ballast) in flight, speed vector,.... I am posting a new thread describing what I just found out also about the NMEA stream from Condor and its limitation - is not providing both date and Height of geoid (MSL) above WGS84 ellipsoid!

By the way, I have been testing the sunset algorithm and find it works
correctly if it is fed the correct GPS date and has the correct timezone
offset for the corresponding location.

Thanks for looking that up and clarifying it.  Paolo Ventafridda/Paul CoolWind was also experiencing some confusion report from XCSoar.  

I hope you don't mind but I thought best to move the posts related to this subject in its own thread.  It was my mistake to write about it as a side note and this post is about using Condor FPL file to prepare XCSoar PRF file.  All the relevant posts were consolidated there too.  It seems that the explanation about Sunset time confusion might not be so esoteric and I think I almost have my mind around it.  See this post: http://www.nabble.com/Need-clarification-about-Sunset-time-calculated-with-Sun-Ephemeris-by-XCSoar%2C-when-used-for-Condor-online-flight-simulation.-td22172716.html

By this I suggest that possibly Condor might always send out today's
date; and secondly that your use of -7 for the timezone corresponding to
Czech is wrong.
[...]

I apologize for the confusion, when I quoted -7 was when I mentioned what was the offset to get local time in my real time zone (PST), US West coast (Las Vegas, Nevada).

Thank again for taking the time to answer my posts.  I am aware that you are very busy with the maintenance and development of XCSoar for the real life pilots community and the odd virtual pilots too.


Best,
Donat

[John Wharington-2]

Hi Donat,

Thanks for your email.  You are not interfering in any way or speaking
out of place; I just wanted to point out that we have to be careful when
catering to Condor users (and I am one!) that we do not introduce
anything that could interfere with XCSoar being used in real life
flying.

Unfortunately I didn't have any time to go flying this weekend as I was
catching up on my day-job work.  Thanks for giving us your background
info, it is nice to hear about people that are part of the "XCSoar
club".

Best regards,


Johnny

On Mon, 2009-02-23 at 15:38 -0800, DPLuigi wrote:

> Hi John,
>
> I hope you had a nice week-end and found some time to fly this week-end.
> You are in Australia, aren't you? I remember vaguely that is was inferred in
> a post somewhere.
>  
>
> John Wharington-2 wrote:
> >
> > Remember that XCSoar is primarily driven by the needs of real-life
> > pilots rather than Condor, and the constraints/drivers from Condor may
> > not be appropriate for XCSoar's primary function.
> >
> I am aware of that and I thought I clearly expressed my concerned to not be
> too off topic.  I thought also I warned at the beginning of my post(s) that
> this was about issues when using XCSoar with Condor.  Anyhow, I hope I am
> now disruptive with my posts on this forum.  Please let me know.
>
> A little about where I come from... I learned to fly for real in the UK,
> during my Ph.D. at UMIST in the UK and I solo-ed there about 14 years ago -
> overall flying for about 2 years.  
> I was fortunate to meet very nice and welcoming members of the local soaring
> club, LVVSA, Jean (Nevada), and was able to fly as a P2 twice since Otcober.
> We all hope I can join as soon as my finance allows me to fly regularly, and
> I would love to re-solo and start where I left off 13 years ago fly
> X-country.
>
> Condor is a very fun tool/flight simulator to keep my mind into it but you
> are right, it is a lame comparison to the real experience I miss very much.
> I am very enthusiastic about finding out how XCSoar helps with the work load
> during XCountry and flight planning.  It is also a sophisticated analytical
> tool that I am learning to enjoy and understand in the process.  I hope to
> be seasoned with it by the time I start X-country flying, which could be
> within a year (or two) and once I solo-ed and flew for a while from the
> "seats of my pants".  In the mean time, Condor is a cheap way to fulfill a
> long standing passion that we all share here.
>
>
>
> > Condor itself is "guilty" of not having everything neatly organised; for
> > example the ballast setting is not saved in the .fpl file.
> >
> I am learning this too while looking with Simon Taylor how to help prepare
> XCSoar configuration from Condor flight plans but also looking into the Free
> Condor Instrument Suite, which suffers form not having access to current
> weight (ballast) in flight, speed vector,.... I am posting a new thread
> describing what I just found out also about the NMEA stream from Condor and
> its limitation - is not providing both date and Height of geoid (MSL) above
> WGS84 ellipsoid!
>
>
>
> > By the way, I have been testing the sunset algorithm and find it works
> > correctly if it is fed the correct GPS date and has the correct timezone
> > offset for the corresponding location.
> >
> Thanks for looking that up and clarifying it.  Paulo Friddaventa/Paul
> CoolWind was also experiencing some confusion report from XCSoar.  
>
> I hope you don't mind but I thought best to move the posts related to this
> subject in its own thread.  It was my mistake to write about it as a side
> note and this post is about using Condor FPL file to prepare XCSoar PRF
> file.  All the relevant posts were consolidated there too.  It seems that
> the explanation about Sunset time confusion might not be so esoteric and I
> think I almost have my mind around it.  See this post:
> http://www.nabble.com/Need-clarification-about-Sunset-time-calculated-with-Sun-Ephemeris-by-XCSoar%2C-when-used-for-Condor-online-flight-simulation.-td22172716.html
>
>
>
> > By this I suggest that possibly Condor might always send out today's
> > date; and secondly that your use of -7 for the timezone corresponding to
> > Czech is wrong.
> > [...]
> >
> I apologize for the confusion, when I quoted -7 was when I mentioned what
> was the offset to get local time in my real time zone (PST), US West coast
> (Las Vegas, Nevada).
>
> Thank again for taking the time to answer my posts.  I am aware that you are
> very busy with the maintenance and development of XCSoar for the real life
> pilots community and the odd virtual pilots too.
> :working:=)
>
> Best,
> Donat
> --
> View this message in context: http://www.nabble.com/Is-it-ok-to-define-Task-and-Task-Rules-from-PRF-file-instead-of-TSK-file--tp22009808p22172729.html
> Sent from the xcsoar-devel mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
> -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
> -Strategies to boost innovation and cut costs with open source participation
> -Receive a $600 discount off the registration fee with the source code: SFAD
> http://p.sf.net/sfu/XcvMzF8H
> _______________________________________________
> Xcsoar-devel mailing list
> Xcsoar-devel@...
> https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

[binky9]

Donat,

When you have converted a Condor task to XCSoar format, where do you save it in XCSoar?  I have XCSoar working with Condor, except the task flight track doesn't show up on the HP310.

Bruce

DPLuigi wrote:

Hi John,

I hope you had a nice week-end and found some time to fly this week-end.  You are in Australia, aren't you? I remember vaguely that is was inferred in a post somewhere.
 

John Wharington-2 wrote:

Remember that XCSoar is primarily driven by the needs of real-life
pilots rather than Condor, and the constraints/drivers from Condor may
not be appropriate for XCSoar's primary function.

I am aware of that and I thought I clearly expressed my concerned to not be too off topic.  I thought also I warned at the beginning of my post(s) that this was about issues when using XCSoar with Condor.  Anyhow, I hope I am now disruptive with my posts on this forum.  Please let me know.

A little about where I come from... I learned to fly for real in the UK, during my Ph.D. at UMIST in the UK and I solo-ed there about 14 years ago - overall flying for about 2 years.  
I was fortunate to meet very nice and welcoming members of the local soaring club, LVVSA, Jean (Nevada), and was able to fly as a P2 twice since Otcober. We all hope I can join as soon as my finance allows me to fly regularly, and I would love to re-solo and start where I left off 13 years ago fly X-country.

Condor is a very fun tool/flight simulator to keep my mind into it but you are right, it is a lame comparison to the real experience I miss very much.  I am very enthusiastic about finding out how XCSoar helps with the work load during XCountry and flight planning.  It is also a sophisticated analytical tool that I am learning to enjoy and understand in the process.  I hope to be seasoned with it by the time I start X-country flying, which could be within a year (or two) and once I solo-ed and flew for a while from the "seats of my pants".  In the mean time, Condor is a cheap way to fulfill a long standing passion that we all share here.

Condor itself is "guilty" of not having everything neatly organized; for
example the ballast setting is not saved in the .fpl file.

I am learning this too while looking with Simon Taylor how to help prepare XCSoar configuration from Condor flight plans but also looking into the Free Condor Instrument Suite, which suffers form not having access to current weight (ballast) in flight, speed vector,.... I am posting a new thread describing what I just found out also about the NMEA stream from Condor and its limitation - is not providing both date and Height of geoid (MSL) above WGS84 ellipsoid!

By the way, I have been testing the sunset algorithm and find it works
correctly if it is fed the correct GPS date and has the correct timezone
offset for the corresponding location.

Thanks for looking that up and clarifying it.  Paolo Ventafridda/Paul CoolWind was also experiencing some confusion report from XCSoar.  

I hope you don't mind but I thought best to move the posts related to this subject in its own thread.  It was my mistake to write about it as a side note and this post is about using Condor FPL file to prepare XCSoar PRF file.  All the relevant posts were consolidated there too.  It seems that the explanation about Sunset time confusion might not be so esoteric and I think I almost have my mind around it.  See this post: http://www.nabble.com/Need-clarification-about-Sunset-time-calculated-with-Sun-Ephemeris-by-XCSoar%2C-when-used-for-Condor-online-flight-simulation.-td22172716.html

By this I suggest that possibly Condor might always send out today's
date; and secondly that your use of -7 for the timezone corresponding to
Czech is wrong.
[...]

I apologize for the confusion, when I quoted -7 was when I mentioned what was the offset to get local time in my real time zone (PST), US West coast (Las Vegas, Nevada).

Thank again for taking the time to answer my posts.  I am aware that you are very busy with the maintenance and development of XCSoar for the real life pilots community and the odd virtual pilots too.


Best,
Donat