On Monday 19 July 2010 03:11:04 am Anders Gidenstam wrote:
> On Sun, 11 Jul 2010, Hal V. Engel wrote:
> > This is a large merge that is a significant enhancement to the existing
> > P-51D but it is still a work in progress and much remains to be done.  At
> > the moment I am not able to spend much time working on this but it is far
> > enough along that I think it belongs in the FG mainline so that it can be
> > tested against 2.1 and hopefully released with 2.1.  I intend to keep
> > working on this as time permits and my intention is to eventually make
> > this the most complete and most accurate model in the FlightGear hanger.
> 
> Hi Hal,
> 
> Lovely aircraft!
> 
> One thing I discovered after a few botched landings is that the crash
> detection system doesn't set /sim/crashed. I think it ought to since that
> is the standard crash indicator in FlightGear. It is easy to add as the
> following diff shows:
> 
> index 312f0ce..bd2e03f 100644
> --- a/Aircraft/p51d/Systems/crash-detect.xml
> +++ b/Aircraft/p51d/Systems/crash-detect.xml
> @@ -56,6 +56,7 @@
>           <output>/sim/freeze/clock</output>
>           <output>/sim/freeze/main</output>
>           <output>/controls/engines/engine[0]/cutoff</output>
> +        <output>/sim/crashed</output>
>        </switch>
> 
>     </channel>
> 
> It is also easy to update the merge request - I have tried it on some of
> mine.

Thanks Anders.  I have made this update and it is now part of my gitorious 
fgdata clone and also part of the merge request.


> 
> PS.
> 
> For those who want to try the p51d before the merge request is accepted
> and aren't that familiar with git here are the steps to get Hal's branch:
> 
> git fetch git://gitorious.org/~hvengel/fg/hvengels-fgdata-p51d-jsbsim.git
>  master:hvengel/p51d-jsbsim git merge hvengel/p51d-jsbsim

I would encourage others to test this since any feedback I get will be used 
for additional improvements.  The model does some unusual things that push the 
limits of JSBSim particularly the propulsion systems and related controls.

There are also some things that don't work correctly because of bugs or 
missing features in the version of JSBSim that is part of FG but that have 
been (presumably) fixed in JSBSim CVS.  These include:

Fuel tank switching is currently broken and the engine will only run on 
tank[0] (left internal wing tank).

The engine gage cluster(s) give wrong oil pressure, oil temperature and 
coolant temperature numbers.  JSBSim has some fixes related to this that do not 
totally fix this but are none the less improvements.

The tail wheel is currently always locked to the rudder.  On the real P-51 
series the tail wheel would swivel if the stick is forward of neutral but lock 
with the stick back.  JSBSim CVS has a fix that will allow for this to be 
correctly modeled and it should only take a fairly simple fcs_function to 
implement. 

Are there any plans to update JSBSim in FG GIT?

> 
> If you have push access to fgdata it might be advisable to create a
> separate branch for this so that you don't accidentally push these
> changes. (As far as I have checked they don't interfere with anything -
> but if we want to keep the gitorious merge request function functional it
> might be better if whoever merges this contribution use that procedure
> instead).

Having this pushed into a branch of the main FG repository for later merging 
into the FG mainline would be fine by me.  This will allow wider testing and 
also allow others to work on this branch.  On the forums I had an exchange 
with another P-51D developer who has his own copy that he is working on.  For 
the most part we are working on different things (he is adding livery support 
and a bombable enabled AI model - I am working on improving the FDM, the 
cockpit, controls and the overall weapons systems) although there is some 
overlap (both have machine guns and are bombable).  The other developer 
expressed an interest in using GIT to get his work into the FG mainline and I 
think his work would be a good adjunct to the update I am proposing.  But it 
was also apparent that he had no idea how to go about doing this.

There was a poll on the forums asking users what they thought was the area in 
FG that needed the most work and by far the item with the most votes was 
improving the existing models.  The more that can be done to encourage 
collaboration on and contributions to the existing models the better it will 
be for dealing with this apparent shortcoming.  It would be really helpful to 
users thinking about working on an existing model to have documentation on how 
to go about this process.  I have worked as a developer on other open source 
projects as well as commercial systems so I had a pretty good idea how to go 
about this but many users do not have this kind of background and have no idea 
how to get started.  

> 
> Btw. local finger trouble with git can easily be adjusted using git
> rebase, see this very useful site for more information:
> http://www-cs-students.stanford.edu/~blynn/gitmagic/ch05.html
> 
> Take care NOT to change published commits, though.
> 
> DS.
> 
> Cheers,
> 
> Anders
> 

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel

Reply via email to