so this weekend I set all curves for the Mk9 spitfire to linear in DCs and BoS, and stuck with it

So my thinking is. there's a FM and there's my personal control curves and they must conflict at some point, with an S curve, yes you get less wobbling around center as the plane is less responsive and then as you start to throw it around it gets more responsive as you get into the steeper part of the curve.

But my, and I'm imaging this no practical flying/FM knowledge but as the plane gets into more extreme maneuvers and higher speeds, there must be more pressure on the CTRL surfaces so less response and more input needed to make them move, at least in a non fly by wire aircraft

more stick input / less FM aircraft movement -> some weird mid ground as your input and FM crossover -> less input / more FM aircraft movement

but with linear input curves, you're only ever flying the FM, not your curves and the FM

stick input-FM aircraft movement -> no input and FM crossover -> stick input-FM aircraft movement

This is the first time i've felt in control, good or bad, and not a passenger suggesting maneuvers and the FM/input working out what they want to do. have only tried it on the Spit in DCs and BF109 in BoS but so far i'm liking this


obviously ignoring FFB sticks that may or may not put up more resistance in a correct way, only have a FFB wheel, but thats linear, and when i feel the front wheels scrub into understeer the load at the wheel goes light compared to full grip on the front.


be interested in your opinion on this

The interface between man and "machine" through the added layer of a simulator can only ever be a compromise (until someone builds the holodeck). With stick input, the major differences start with the center of rotation / pivot point already. Makes a huge difference as to the granularity of control.

I remember the first time I went up in a glider iRL and thinking "Man, this FM seems unrealistic" haha. Because it was very stable and felt like what some refer to as "flying on rails" in a flightsim. The reason for that of course was - other than the much higher play of the stick - that gliders are naturally very stable a/c because of their HUGE wings and slim bodies etc

Another time I went up in a "Sunwheel" https://www.youtube.com/watch?v=ZxxtgbOVqLc (vid not from me, but that's the plane), and this a/c felt VERY twitchy. Cudos to the devs of Rise of Flight here, their FM feels very close to that, especially how light on the roll axis they are.

Rather than "granularity", I should have maybe said "resolution". But I think you get what I mean.

And to be more specific to your posting: Whatever works for you. I don't have the Spit, but I got almost all the other planes including the Mustang and Dora. For "AFM" planes I always make the curves fully linear and never had problem. Only with the SFM planes do you get much too aggressive reaction on small stick inputs / not enough "fine control" from my experience. Probably because the realistic FM with AFM planes already puts in a sort of natural damper because of the inherit momentum / inertial mass of the plane and airflow simulation or whatever else they did. Still, if someone has a stick with only little play, he might want to input some curvature to get the fine control back. The problem with curves is the spot where it transitions from fine to "accelerated", which can easily lead to over-input and stalling the plane in a turnfight for example.

I used to use quite a bit of curve, in some aircraft, and less in others. However, adding 12.5 cm (5 in) of extension to my Warthog made a world of difference. If I use any curve at all, now, it is very little but most aircraft are flown with no curve at all. The longer stick makes small movements so much easier.

awesome thanks guys

yea i have an extended warthog too, so i felt like i was over correcting everytime i move, but now im more conscious of what the plane is doing with my input and i feel i'm, for good or bad, flying the FM

BTW, while I understand what you mean, I would like to point out that you are always "flying the FM". The point is how well the control system in itself is modelled on the one hand and how far you on the other have to make compromises to account for the differences between a real world Spitfire stick and a "Joystick" (which you already mitigated to a large extend by increasing the distance to the pivot point). Even in the real world you are always only interfacing with the control surfaces and - in extensio - the airflow, be it through only cords, or a hydraulic control system, or a stability augmentation system on top of that, or a Fly By Wire system on top of that.

I found an interesting video a few days ago that somewhat pertains to this discussion and you might find interesting, too:

https://www.youtube.com/watch?v=w_hAVknWzzQ

It explains in detail the flight control and stability augmentation system of the F-4 Phantom. I have the Milviz Phantom for FSX and I find its SAS is very nicely simulated. You can disable / enable the SAS for all three axes (pitch, roll, yaw) seperately, like in the real bird. When you do a roll with the SAS engaged, the moment you center the stick, the SAS will apply some short opposite aileron input to try and stop the roll instantanously, while with it disabled, it will continue to roll a bit because of momentum. Like it is said in the video though, SAS enabled might not always be preferable, since when rolling in on a target / lining up for a gunpass etc, you might want to have some more natural smoothing in of the jet rather than it trying to get stuck in the "commanded" position and you hamfisting it to where you want it to go.

Another very extreme example for flight control systems is with the F-104 Starfighter. Because it has so such tiny wings, it actually has what is called a "Boundary Layer Control System". That system actually blows Bleed Air from the turbine over the wing surfaces when at low speed (when flaps are extended) to add much needed airflow that would be too little if the wings were left to their own devices. So, the point is that in a Starfighter you are flying the final right down to touchdown with actually 75% throttle at least, or you WILL stall out. There is a very realistic Starfighter for FSX from Sim Skunk Works that actually simulates that system. The first few times I flew it, I didn't know about that peculiarity and wondered why the heck I was stalling on final even when still at 180 knots IAS lol! I was wondering if I was really supposed to come in at 220 knots or something? Then I read the real life manual and Skunk Works manual of the Starfighter and learned about that system. The mistake was that I was near Idle or at Idle on final, while you have to be somewhere around 75 to 85 % throttle, and the full flaps and if necessary speedbrakes (and landing gear) will be enough to keep your speed down, while that power setting will provide the "artificial" airflow for the wings not to depart. Crazy!

I have the F4 ADV, I'll do those SAS on/off trials next time

Good video

I think its the weirdness as my fake curves cross over into sensitive, i might try a reverse S. so at high G full stick i have less direct control to simulate ctrl surfaces being "heavy"

I don't even have the Advanced Version, just the "standard" (find it pretty rough to ask extra for what imho is the completion of the plane, especially since I bought it way back when it first came out in beta and there wasn't a word on paying extra for several degrees of completion - it was marketed as a "hardcore" version from day one). Luckily the SAS is already simulated in the standard version.

Re the curves again - I feel like the more you play around with the curves in AFM planes, the more weird the result might get. I think you cannot really simulate control forces by changing stick effectiveness. First order of the day should be to get good control. Stick forces you can probably only reasonably simulate via actual force feedback. Control forces ≠ Stick response.

yeah i had the beta as well, totally agree about tacpac addon and now adv ...but im a sucker for Double Ugly and thats the best version out there

im going to carry on with linear curves, as to my mind it takes one variable out of the equation and ill see how i get on. very quicly linearised my F-5 dcs curves last night and its night and day difference from the spit 9 as expected but also a different plane to lfy than with curves,

im going to look into modifying my stick setup with external resistance springs