homepage Forums Air Combat & Civil Aviation Battle of Britain + Battle of Britain II Reduce control input

I know this is a controversial topic, but I think it should be discussed again.
I think the planes in BoB are too twitchy. I have learned to live with it and can dogfight at low altitudes without entering a spin too often. But not all that makes a sim more difficult necessarily makes it more realistic. Plus it keeps newcomers out that are otherwise interested. Right now I am trying to drag in a friend who is complaining that the response to the tiniest input is too drastic even in "sim" mode and I know what he is talking about.

I am not asking to change the entire flight model. But right now the relevant control inputs are concentrated in a concentric ring around zero with an amplitude of say 5 per cent of maximum stick deflection. Any input above this will send you into a spin immediately. Why not reduce the control surface deflection for the same stick travel to make more precise control feasible? This would increase the non-lethal stick movement to say 10 percent. Don't go by the numbers, they were made up to get the idea across.

We would not be giving away anything. Even IL2 allows every player to customize (and limit) the response of his joystick and this sim isn't exactly known to sacrifice realism for ease of use.

Prion

Prion !
Good point. I assume you have been playing around with "dead-zone". Are you using FFB ?
I'm using the gameport on my SB "Live".
Happy Landings
Benny

I second the motion.

This has been discussed before, and I assume it is on somebody's list. Great advancements have been made, but this is one area that has not been addressed, and I consider it a "bug" from the original game.

I use a USB FFB stick and I am not exagerating when I say that just the feedback judder from firing the guns makes the nose of the plane dance all over the place! I have tried widening the dead zone but this is counter productive because then you have to anticipate the edge of the zone where all the control is concentrated and slow the stick deflection rapidly. This is not easy.

A user adjustable stick input scale factor would be my preferred solution.

The planes do not exactly greet the newcomer do they?

It has taken me a long time to really get a feel to them. And even now I feel a rank amateur.

Mind you yesterday I had two fantastic dog fights, and a PERFECT landing. It was such a good session I immediatly knew I would have brag to the giys here-sad ain't it!

For once I did not use the outside view (which I still find necessary)and it was so smooth that I honestly did not know I was down! And the dog fight was at on stage so low if I had been able to extend the u/g it would have touched the ground.

So maybe the secret is just practice practice practice??


The Cat

As I said above, realism and difficulty are related, but they are not the same thing. I am not asking to dumb down the flight model to make the sim more accessible to newbies (which we all were at some stage BTW).
I am asking to reduce the oversensitivity of the stick control input, the flight model be untouched. The only thing that would change is that more stick travel will be needed to achieve the same control surface deflection in the sim. Thus reducing the occurrence of unwanted maneuvers and spins. Mind you, I am talking about good (sim) pilots, the proverbial ham-fisted ego-shooter pilot type would send his plane into a spin just the same. He just needs more intentional stick movement for it. When flying a real plane, there are so many more cues whether your plane is going to stall soon that it seems hard to ignore.
It is indeed possible to get used to the controls just happening with the oh so tiniest movements of the joystick, but this does not simulate the difficulties of flying a real airplane IMO. So why not giving a bit more stick travel for more precise control, or rather: Let's keep flying the plane as difficult as it is now, just make it easier to tell the joystick which movements on the stick we actually intend.

Prion

Prion, I agree that the flight model is good and that the stick input is hyper sensitive. Some sticks seem to be more smooth than others. The stickworks tool helps, but doesn't seem to solve the problem completely. I am far from an expert, but there might also be an issue with the control surfaces. If i go to outside view, i notice that there is a certain amount of twitch that occurs to the control surfaces with no stick input, particularly the rudder. If this is the case, then even making the stick less sensitive would not achieve the required effect. But not being an expert at any of these things, I would yield to more informed sources.

I do realize that this is the main problem of some people. The reason we have not done the suggested fix is that it is in my humble opinion a kludge (in German "Pfusch") . From comparing what people have told me and what I see myself with my el-cheapo joystick, I think that there is something fishy going oin in ther joystick input routines. Probably they do something (allowed or unallowed) exotic that some sticks do not like. The reason I say this is that a) for some people the input jidders without touching the stick and the sensitivity seems very different for different people. So it is not that Rowan just implemented a too high sensitivity, that would explain neither of those two effects. Of course it can be that additional to the exotic code there is a sensitive input 8this is quite probable, since for example the Spitfire was extremely sensitive in pitch), but the correct way to work would be to investigate and solve the two effects first and then look at sensitivity.

Anyway, the people with probelms can make some experiments that may help me:

0. Write here your specs (CPU, RAM, OS, Joystick, possibly Joystick drivers, possibly joystick tweakers like CTJF). Also, after each oif the following steps, write the results here.
1. Test whether without touching the controlls, the ailerons, elevator and rudder and throttle move. If there is no movement, you are done.
2. Go into windows joystick calibration and see how it behaves without touching it.
3. Exit BoB, open bdg.txt in an ascii editor, search for REINIT_JOYSTICK_OFTEN. Is it ON or OFF? switch to the other setting and redo tests 1 and 2.

Osram,
I see your point and it is quite honourable to weed out the cause of the problem right away. But meanwhile it would help people to be able to limit the response of their joysticks, i.e. equal 100% of joystick travel 80% of the maximal input or whatever they please. IL2 does it, Janes F18 does it, so why not BoB?
If not all users need the same magnitude of reduction, fine, they will use what values they need then. And even if this does not cure the root of the problem, it would certainly improve their situation. Nothing would be lost as setting 100/100 would keep everything as it is now. But it would be one handle to experiment with rather than none, until the underlying problem is properly fixed.
Don't get me wrong Osram, your way is certainly preferred, I am just trying to get more people in and this is their main complaint: How can I make the joystick less responsive?
I know my flying in IL2 began to match more closely my intentions when the utility to downscale joystick response was added in a patch.

Prion

PS: my joystick is digital and shows zero jitter, just way too sensitive
XP1700, TM Cougar, Voodoo5, Win2k with SP3, DX8.0a

Hi.
I do not want to get in the middle of an old and heated discussion but I have been looking at the code today. There might be solution that satisifies both sides (I am not totally sure).

The way the "deadzone" is impemented needs some analysis. It is almost like none then all on the edges. They are using a quadratic to model the deadzone which is very steep on the ends. We might want to look at something at alot slower. This would smooth out going from deadzone to no deadzone.The folks with no deadzone will not be effected but the folks with the deadzone will have a smoother transition at its edge.

Osram and Retro any comments on this possible approach?
Buddye

Osram, hope this helps:
900Mhz PentiumIII
384MB Ram
Windows ME
Logitech Wingman Extreme with latest drivers
CTFJ installed
Radeon 7000 video card

initial test ;
visually,
aileron and elevators steady,
throttle and rudder jiddering quite markedly.
calibration screen shows same observation.

in bdg.txt, turned REINIT_JOYSTICK_OFTEN from ON to OFF.

second test;
aileron and elevator steady,
throttle and rudder jiddering but apparently much less so.
calibration screen shows aileron and elevator steady, throttle and rudder jiddering slightly.

Probably a different problem, but rudder, after being applied has trouble centreing despite endless tweaking. Also the initial application of rudder will often operate with reversed polarity, but after being fully extended several times in both directions, the polarity will correct itself. (maybe i should use my joystick for a football )

Osram and Buddye1,

Thanks for looking into this

My specs:
OS: Win98SE
CPU: P3 1.2
Joystick: Microsoft ForceFeedback2 USB Stick
Driver: This stick has its own drivers that replace the windows joystick routine.

Init Joystick Often = ON; No jitters of controls
Init Joystick Often = OFF; No jitters of controls

If you can, please consider making the x axis, y axis and z axis independently adjustable. Pitch seems much worse than roll for me. Rudder seems about right.

Hi
I tried a different equation (linear)that was already built into the program and to me (a BoB beginner) it was better. Very smooth in the edges of the Deadzone but nearing the ends of the stick still very sensitive.
Buddye

Hi, short update
The equation must be better and smoother as I can now take off in a Spit and I never could before on the realistic settings.

Speaking of very sensitive. With Airframe Stress equal on (in preferrences), you just have to move the stick up and down alittle to break the wings. Anyone else noticed this........
Buddye

Budye1 !
"Speaking of very sensitive. With Airframe Stress equal on (in preferrences), you just have to move the stick up and down a little to break the wings. Anyone else noticed this........"
This has been up before, and it's wrong. You can apply way too much "G", too quickly.
Happy Landings
Benny

you just have to move the stick up and down alittle to break the wings. Anyone else noticed this........

Funny you should note that Budd, I turned right slightly and pulled up gently while on finals in sight-seeing (no stress) and the right wingtip broke off Fullest realest settings

I thought it might be me though as I'm coming back from what turned out to be premature retirement

Ming

I leave airframe stress OFF for that reason. Clipping one's wings by just looking at the joystick doesn't sound full real to me
Prion

Thanks for the feedback. Like I said, I do think this is an important area.

Buddye1, sounds very interesting!

About the wing breaking, I got this exactly once. Again, I am at fairly much of a loss as to why it is so different for different people.

And it is true that one issue with the current code is that you can move the control surfaces much too fast. We will have to implement a limiter that limits how fast you can move the surfaces .

Buddye1, I wonder...is that equation you mention something to do with the alternative to "Realistic" - the "Novice" setting, which is noticeably smoother? Does that routine get called when Novice is chosen, instead of the hairier Realistic job?

S!

With the help from the Thrustmaster support I found out how to limit the response of my joystick
I am afraid that this will only work if you are using the same joystick model (TM Cougar)
http://forums.frugalsworld.com/vbb/showthread.php?s=&postid=343782#post343782

But it is a totally different flying now, and it feels much more believable to me. The plane is as responsive as ever, I just make slightly larger joystick movements to fly it.

I wasn't aware that this was different for every user, actually. I am very much looking forward to a BoB specific solution. The results on my behalf are very encouraging. BoB can be made, well not exactly easier, but more intentional, if that is a word, without sacrificing realism at all.
Prion

Hi Sneaky, No I am talking Realistic only. Actually, I found a bug and both Realistic and Novice were being sent through the same applydeadzone program. I just used another existing option to apply the deadzone (very simple change). I am using deadzone "large" as a setting and to me its much smoother at the ends of the deadzone. I will talk to Osram to be sure but we may need to get some other users to test it.
Buddye