Hmm... not sure if I understand you correctly here, but if you use <Ts>0.05</Ts> it should mean that your controllers will be clamped to 20Hz regardless of the actual frame-rate.
However, there are pros & cons to doing this. On the plus side the controllers should work consistently at 20fps and greater but on the down side it means that you are defaulting to a lower controller resolution and you may find it harder to tune the controllers. A technique I've recently been experimenting with, and which is providing good results, is to limit the output range of the controller and then restore the full range using a simple gain filter. I'm not sure about exactly how this works but, like I say, it has been effective here. The specific examples I've used have been for pitch and roll controllers where the controllers output a flight-surface deflection i.e. elevator-trim or roll-trim against a target pitch or roll. Normally for these the controller outputs would be limited to the deflection norms i.e. +/- 1.0 (<u_min>-1.0</u_min> & <u_max>1.0</u_max>) but what I've done is to reduce this to +/- 0.25 (<u_min>-0.25</u_min> & <u_max>0.25</u_max>) and then multiply the output stream by 4.0 to regain the full +/- 1.0 normalised range. Have a read of the new README.digitalfilters doc to see how you can also use a reciprocal filter to reduce the gain of a PID controller as airspeed (or mach) increases, so that a controller that works ok at relatively low speeds can be prevented from going into oscillation at higher speeds (this can also be used to reduce stick sensitivity with speed). Using these techniques I've been getting some good results with controllers that work ok here between ~20 & ~60+ fps over wide airspeed ranges. LeeE On Friday 22 February 2008 12:24, flying.toaster wrote: > Thanks for the advice > > I have tried that and changing derivation time but it did not > work... I use the speed-throttle to emulate low FPS on my system > though so I do not not how good my tests were. If anybody with > 20fps is willing to try this fix, please let me know if it is > successful > > Regards > > Enrique > > > Message du 21/02/08 à 16h17 > > De : "LeeE" <[EMAIL PROTECTED]> > > A : [EMAIL PROTECTED], "FlightGear developers > > discussions" <flightgear-devel@lists.sourceforge.net> Copie à : > > Objet : Re: [Flightgear-devel] Announcement : F-14 early alpha > > on CVS > > > > Use the <Ts> tag in the controller configs to specify a 'fixed' > > sampling rate. Personally, I config and tune all my > > controllers to run at 20Hz so that there is a good chance > > they'll work on lower-end systems without causing problems on > > faster machines. > > > > LeeE > > > > On Wednesday 20 February 2008 21:56, flying.toaster wrote: > > > There seems to be a problem with the PID controller > > > implementing the pitch SAS when FPS is low (around 20fps). > > > Probably an interaction between sampling rate and dynamics. > > > People with high end computers will not notice it. I'll try > > > to develop a fix (write the SAS PID in the Nasal code and > > > clean it up) > > > > > > Sorry guys ! > > > > > > > Message du 20/02/08 à 21h50 > > > > De : "flying.toaster" <[EMAIL PROTECTED]> > > > > A : "flightgear-devel" > > > > <flightgear-devel@lists.sourceforge.net> Copie à : > > > > Objet : [Flightgear-devel] Announcement : F-14 early alpha > > > > on CVS > > > > > > > > Hello, > > > > > > > > An early alpha of the F-14B has been kindly put on the CVS > > > > by Alexis for testing. > > > > > > > > Beware of the following limitations : > > > > - There is no cockpit, only a HUD > > > > - Nasal code needs cleaning and optimisation > > > > - Some textures are missing, some animations are incomplete > > > > - Key bindings are non standard > > > > - Early testing on some configuration exhibit instability > > > > on the FDM > > > > > > > > This release is only meant to get early feedback on the > > > > flight model (other suggestions for improvement can be > > > > considered though). When giving feedback (through direct > > > > e-mail or the forum) please consider describing the > > > > configuration of your computer so that I can get a better > > > > idea of what may be the cause. > > > > > > > > If you feel like testing a very broken aircraft ... enjoy > > > > ... But please consider giving feedback to help improving > > > > it > > > > > > > > Cheers > > > > > > > > Enrique > > > > > > > > > > > > ----------------------------------------------------------- > > > >---- ---------- This SF.net email is sponsored by: Microsoft > > > > Defy all challenges. Microsoft(R) Visual Studio 2008. > > > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > > > _______________________________________________ > > > > Flightgear-devel mailing list > > > > Flightgear-devel@lists.sourceforge.net > > > > https://lists.sourceforge.net/lists/listinfo/flightgear-dev > > > >el > > > > > > ------------------------------------------------------------- > > >---- -------- This SF.net email is sponsored by: Microsoft > > > Defy all challenges. Microsoft(R) Visual Studio 2008. > > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > > _______________________________________________ > > > Flightgear-devel mailing list > > > Flightgear-devel@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/flightgear-devel > > > > --------------------------------------------------------------- > >---------- This SF.net email is sponsored by: Microsoft > > Defy all challenges. Microsoft(R) Visual Studio 2008. > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > _______________________________________________ > > Flightgear-devel mailing list > > Flightgear-devel@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/flightgear-devel > > ----------------------------------------------------------------- >-------- This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Flightgear-devel mailing list > Flightgear-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/flightgear-devel ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
