Jump to content
The simFlight Network Forums

Recommended Posts

Posted

Hello there,

Something's not quite right. For some reason, calibration is not being applied properly (after starting my computer anew). I have a Logitech Force 3D Pro since a couple of weeks, it needs power in addition to a USB connection. I do not always have its power on and so the joystick is not "connected" to my computer all the time, I think. I also have the Logitech Gaming Software (LGS) installed and it's working fine as far as I can tell. There's one option I'm not sure about, "Switch selected device to ID 1", though my problem seems to happen with and without the option enabled.

It goes kinda like this. I start my computer, log on, turn on the power of the joystick and start the LGS. Everything fine so far. Then I start FS9. Its sensitivity and dead zone options are set to simplified (I think it's called, so same setting for all axes), former set to about 50%, the latter to 100%. I am aware of the user guide recommendations, but these settings seem to work best for this joystick. In other words, this gives a pretty good result in the sim already. I use FSUIPC (3.966, though my problem is not unique to this version) to then finetune it all. When FS9 is running, I switch to slew mode and calibrate on the slew tab of the calibration section. (I make a lot of precise slew movements when I'm editing Afcads, so slew needs to be calibrated properly.) Calibration works fine (I do not calibrate altitude change), add a slope of 5 to the other three, click on OK, everything works exactly as I like it! I then close FS9 (end flight, then exit) and shut down my computer.

Then I start it up again and check both FS9 and FSUIPC config file to verify the settings, they are all there and correct as per my actions above. I don't make any changes. I repeat the same steps to get the sim running, switch to slew mode and then the problems start, If I as much as point or blow at my joystick, the default Cessna starts moving. If I twist it either way, the Cessna starts spinning like crazy. Same with the other movements, it moves great distances with only gentle movements of the joystick. Why is this happening? Everything was fine when I left FS, the settings/numbers are there in the config files, yet the joystick is out of control...

I've read through the user guide about calibration, I thought that maybe setting AutoAssignLetters to Yes would help, no luck. There are no other game devices on this PC. I'm currently using Windows 7, 32-bit and it was a clean install, about three weeks ago. Still "clean" today, no other issues, in fact, it has been a positive experience coming from XP.

Now, when I close FS9 and then restart it without shutting down my computer, calibration seems to stick. If I close FS9, disconnect the power of the joystick, then reconnect and start FS, calibration seems to remain. But again, shut down my computer, start it up again, FS9, crazy joystick...

Below are relevant settings from FS9.cfg:

[JOYSTICK_MAIN {9FC5FA50-0906-11DF-8001-444553540000}]

AXIS_EVENT_00=AXIS_AILERONS_SET

AXIS_SCALE_00=63

AXIS_NULL_00=127

AXIS_EVENT_01=AXIS_ELEVATOR_SET

AXIS_SCALE_01=63

AXIS_NULL_01=127

AXIS_EVENT_05=AXIS_RUDDER_SET

AXIS_SCALE_05=63

AXIS_NULL_05=127

AXIS_EVENT_06=AXIS_THROTTLE_SET

AXIS_SCALE_06=63

AXIS_NULL_06=127

BUTTON_DOWN_EVENT_00=BRAKES

BUTTON_DOWN_REPEAT_00=1

BUTTON_DOWN_EVENT_01=VIEW_MODE

BUTTON_DOWN_EVENT_02=ELEV_TRIM_DN

BUTTON_DOWN_REPEAT_02=1

BUTTON_DOWN_EVENT_03=ELEV_TRIM_UP

BUTTON_DOWN_REPEAT_03=1

BUTTON_DOWN_EVENT_04=FLAPS_INCR

BUTTON_DOWN_EVENT_05=FLAPS_DECR

BUTTON_DOWN_EVENT_06=KNEEBOARD_VIEW

BUTTON_DOWN_EVENT_07=GEAR_TOGGLE

BUTTON_DOWN_EVENT_08=VIEW_MODE

BUTTON_DOWN_EVENT_09=GEAR_TOGGLE

POV_MOVE_EVENT_00=PAN_VIEW

POV_MOVE_REPEAT_00=1

[JOYSTICK_SLEW {9FC5FA50-0906-11DF-8001-444553540000}]

AXIS_EVENT_00=AXIS_SLEW_SIDEWAYS_SET

AXIS_SCALE_00=63

AXIS_NULL_00=127

AXIS_EVENT_01=AXIS_SLEW_AHEAD_SET

AXIS_SCALE_01=63

AXIS_NULL_01=127

AXIS_EVENT_05=AXIS_SLEW_HEADING_SET

AXIS_SCALE_05=63

AXIS_NULL_05=127

BUTTON_DOWN_EVENT_00=SLEW_RESET

BUTTON_DOWN_EVENT_01=VIEW_MODE

BUTTON_DOWN_EVENT_02=SLEW_ALTIT_PLUS

BUTTON_DOWN_EVENT_03=SLEW_ALTIT_MINUS

BUTTON_DOWN_EVENT_04=SLEW_PITCH_PLUS

BUTTON_DOWN_EVENT_05=SLEW_PITCH_MINUS

BUTTON_DOWN_EVENT_06=SLEW_BANK_MINUS

BUTTON_DOWN_EVENT_07=SLEW_BANK_PLUS

BUTTON_DOWN_EVENT_08=SLEW_FREEZE

BUTTON_DOWN_EVENT_09=SLEW_RESET

POV_MOVE_EVENT_00=PAN_VIEW

POV_MOVE_REPEAT_00=1

And from and FSUIPC.ini:

[JoyNames]

AutoAssignLetters=Yes

0=Logitech Force 3D Pro USB

A=Logitech Force 3D Pro USB

[JoystickCalibration]

AllowSuppressForPFCquad=Yes

ExcludeThrottleSet=Yes

ExcludeMixtureSet=Yes

ExcludePropPitchSet=Yes

SepRevsJetsOnly=No

ApplyHeloTrim=No

UseAxisControlsForNRZ=No

FlapsSetControl=0

FlapDetents=No

ReverserControl=66292

Reverser1Control=66422

Reverser2Control=66425

Reverser3Control=66428

Reverser4Control=66431

MaxThrottleForReverser=256

AileronTrimControl=0

RudderTrimControl=0

CowlFlaps1Control=0

CowlFlaps2Control=0

CowlFlaps3Control=0

CowlFlaps4Control=0

MaxSteerSpeed=60

SlewSide=-4064,-512,512,4063

SlewAhead=-4064,-512,512,4063

SlewHeading=-4064,-512,512,4063

SlopeSlewSide=5

SlopeSlewAhead=5

SlopeSlewHeading=5

Would setting STICK_SENSITIVITY_MODE to 0 help? Not sure what the unwanted behavior is that the user guide mentions. I've also tried calibrating regular flight mode and that seems to work, but I've not tested it fully, I'm trying to concentrate on slew first. Values for regular calibration go from about -16000 to +16000, except for the throttle, which goes from about -4000 to +4000, like the slew values, not sure if those values are normal.

Finally, I'd like to mention ActiveCamera 2004, specifically a problem I had with this joystick initially, the following is taken from the AC site's FAQ (http://www.anticyclone.be/faq.htm)

Q: when I turn off the computer I get the closing program for FS9. In other words it has not closed when I exit Flight Sim some hours before.

R: Due to a conflict with DirectInput on some computers the FS9.exe process is still running after FS 2004 is shut down. Therefore it was not possible to save the Active Camera settings in the setup menu.

Download the fix (see the Download page at http://www.anticyclone.be). This dll fixes the issue. Note that you will no longer be able to assign Active Camera events to any joystick button.

My FS9.exe did get stuck, I applied the update, problem gone. But perhaps some issue still remains?! Who knows anymore... Sorry for the long post and perhaps I'm missing something, but I'm all out of ideas, big headache though.

:lol:

You can say a lot about Microsoft, but my previous Microsoft SideWinder Precision 2 joystick never gave me any trouble!

Posted
Then I start FS9. Its sensitivity and dead zone options are set to simplified (I think it's called, so same setting for all axes), former set to about 50%, the latter to 100%. I am aware of the user guide recommendations, but these settings seem to work best for this joystick.

100% dead zone? That would mean no live part at all! You cannot mean that.

And 50% sensitivity might be okay if you aren't calibrating in FSUIPC, but if you are then it's a complete waste of your device's capabilities. For FSUIPC calibration you either want to assign in FSUIPC, or you want full sensitivity (100%, or full right) and no dead zone (full left, zero).

Then I start it up again and check both FS9 and FSUIPC config file to verify the settings, they are all there and correct as per my actions above. I don't make any changes. I repeat the same steps to get the sim running, switch to slew mode and then the problems start, If I as much as point or blow at my joystick, the default Cessna starts moving. If I twist it either way, the Cessna starts spinning like crazy. Same with the other movements, it moves great distances with only gentle movements of the joystick. Why is this happening? Everything was fine when I left FS, the settings/numbers are there in the config files, yet the joystick is out of control...

Something is obviously changing in between.

I've read through the user guide about calibration, I thought that maybe setting AutoAssignLetters to Yes would help, no luck

Sorry, but there's nothing in anything I've ever written which would explain such erratic and completely unpredictable behaviour. Something is evidently changing in the route between the joystick and where it gets to FSUIPC -- check the "INPUT" numbers, see the differences.

Now, when I close FS9 and then restart it without shutting down my computer, calibration seems to stick. If I close FS9, disconnect the power of the joystick, then reconnect and start FS, calibration seems to remain. But again, shut down my computer, start it up again, FS9, crazy joystick...

Ouch! It is starting to sound like some sort of driver or USB problem on your PC.

Below are relevant settings from FS9.cfg:

Well that's not at all relevant to me, except that you most certainly should have zero NULL and maximum sensitivity. If you don't, then please don't bother trying to calibrate in FSUIPC. There's no point. It is all designed to operate correctly only on the full undisturbed range of your controls. I think you are undermining it somewhat.

The other thing which may also be a factor is the way FS handles joystick inputs. It was changed after FS98 from being a linear response to a time-based response. In my opinion that was a bad move, and combined with messing so badly with the numbers by having wrong sensitivity and null zones I think you could end up with a mess in any case. You could try doing as advised in the FSUIPC user guide, setting Stick Sensitivity Mode to 0 (this reverts FS's treatment to pre-FS2000 methods, more linear). But I still can't vouch for what will happen with your odd null and sensitivity settings which will tend to screw up the numbers in any case (they operate as subtractors and divisors, respectively).

And from and FSUIPC.ini:

I notice from that that you've not even bothered to calibrate your controls in FSUIPC, only the slew axes. Have you never even tried to use FSUIPC for its truly intended purpose in this area? (Oh, I see you have, later in your message. Odd you should have then deleted those).

Would setting STICK_SENSITIVITY_MODE to 0 help? Not sure what the unwanted behavior is that the user guide mentions.

It's a long time ago now (10 years) but I think it otherwise exhibits non-linear behaviour based on the speed you change the axis. Faster movements = greater change, I think. I've never delved into it that much, it was noticeably a bad idea from the get-go.

Have you thought of assigning in FSUIPC in any case, and bypassing all that unpredictability?

Regards

Pete

Posted
100% dead zone? That would mean no live part at all! You cannot mean that.

I didn't say 100% dead zone, I said the option is set to 100% or full right as you put it...

Something is obviously changing in between.

Obviously...

Sorry, but there's nothing in anything I've ever written which would explain such erratic and completely unpredictable behaviour.

No, but I never said such a thing... I thought that maybe the joystick was being incorrectly identified after a boot. Now perhaps I misunderstood the User Guide, but I thought that having a letter assigned, either manually or with the auto option, actually decreases the chance of incorrect identification.

Something is evidently changing in the route between the joystick and where it gets to FSUIPC -- check the "INPUT" numbers, see the differences.

As in the values I list later in my post? What differences? Between IN and OUT?

Well that's not at all relevant to me

Yeah, what do joystick settings from the FS9.cfg file have to do with joystick calibration? I did my best to provide as much information as possible from the start, so you wouldn't have to ask me a bunch of questions. If some of that information is of no use to you, fine, but you don't have to rub it in, nor put words in my mouth. I find your tone unnecessarily unpleasant and instead of getting of the bottom of this, I now find myself refuting you, which doesn't help either of us.

except that you most certainly should have zero NULL and maximum sensitivity. If you don't, then please don't bother trying to calibrate in FSUIPC. There's no point. It is all designed to operate correctly only on the full undisturbed range of your controls. I think you are undermining it somewhat.

I quote from the User Guide:

"Go to the FS joystick options and make sure you have the sensitivity at maximum and the null zone at minimum. Any other setting will reduce the range of your joystick and make FSUIPC‘s more accurate calibration less effective."

"Most certainly should"? "There's no point"? "Operate correctly only"? Or "undermining it somewhat"? "Reduce the range"? "Less effective"? Let me get one thing straight. What is it? Is it absolutely one hundred percent guaranteed only to work with max sensitivy and zero null? Or should it work, merely diminishing the effectiveness somewhat? And how does that relate to it working for me one moment, with those settings, but not the other?

I notice from that that you've not even bothered to calibrate your controls in FSUIPC, only the slew axes. Have you never even tried to use FSUIPC for its truly intended purpose in this area? (Oh, I see you have, later in your message. Odd you should have then deleted those).

I have deleted items from the ini file yes, to make sure everything was reset for another try. It may just have been my imagination, but using the reset buttons at one point, didn't reset the slope values. Calibration was reset, but the slope values were still there. Perhaps those slope values are only active with calibration values? Or they are reset when new calibration starts?

Have you thought of assigning in FSUIPC in any case, and bypassing all that unpredictability?

Yes, but it seems like overcomplicating things to me. Apart from the axes, I have also considered using FSUIPC to program my buttons, but the Logitech Gaming Software (LGS) is handling that perfectly now and it has a neat export/import tool. And my Force 3D Pro is a Force Feedback joystick to boot, something not supported by FSUIPC.

What I'll do after lunch is the following, set sensitivity to max and null to min, and use the STICK_SENSITIVITY_MODE=0 setting. We'll see what happens.

Posted

I didn't say 100% dead zone, I said the option is set to 100% or full right as you put it...

Well, maximum dead zone -- it cuts out a huge portion of the axis capability.

I thought that maybe the joystick was being incorrectly identified after a boot. Now perhaps I misunderstood the User Guide, but I thought that having a letter assigned, either manually or with the auto option, actually decreases the chance of incorrect identification.

The lettering of axes is only relevant to axes assigned in FSUIPC. In your case you only appear to have one device in any case:

0=Logitech Force 3D Pro USB

A=Logitech Force 3D Pro USB

When you assign, in FSUIPC, to an axis on Joystick "A", FSUIPC finds that in the list above, then uses the name to find the Windows ID number, 0 in this case. But with only one device even if without this, if things did get mixed up, it would only result in no assignments working at all -- because the joystick wouldn't be on the ID number programmed into FSUIPC.

Since you are assigning only in FS, none of this is at all relevant. FSUIPC is not reading the joystick, FS is. All the calibration in FSUIPC does is calibrate the internal FS controls, not the joystick itself but the results after passing through FS's assignments and manipulations.

What can happen, if you disconnect and reconnect devices which were assigned in FS is that the assignments are re-made, affresh, because FS can assume the device is new. I think this is because the Windows registry entries for the device get re-built, and FS uses some strange "GUID" number to match things up which might change in the process of Windows continually re-installing a USB device. The GUID in the FS CFG extract you showed is this:

{9FC5FA50-0906-11DF-8001-444553540000}

See if that changes at all. That could certainly mess things about.

Yeah, what do joystick settings from the FS9.cfg file have to do with joystick calibration?

No need to be sarcastic. FSUIPC calibration has nothing whatsoever to do with FS's CFG file. FSUIPC calibration works on the internal FS controls. It knows nothing about joysticks, assignments, whatever. That was all I was saying.

If you use FSUIPC assignment, I can tell you exactly what is going on, because then FSUIPC is reading the joysticks. I cannot help with what FS might do when you assign there, except my advice already given and made clear in the documentation -- zero dead zone, max sensitivity, Sensitivity Mode zero.

Is it absolutely one hundred percent guaranteed only to work with max sensitivy and zero null? Or should it work, merely diminishing the effectiveness somewhat?

It really makes it less than worthwhile. It makes what it is trying to do -- i.e. make the whole range of your axis usable for the whole range of FS's capabilities -- a bit futile.

And how does that relate to it working for me one moment, with those settings, but not the other?

It does not at all, as I thought I said.

using the reset buttons at one point, didn't reset the slope values. Calibration was reset, but the slope values were still there.

Erthe "Slope" button won't be there after a Reset because Reset cancels the calibration altogether. Do you mean after pressing "Set" again after "Reset", to re-enable the calibration? I'll have a look to see whether the Slope should be reset to 0 or not.

[LATER] Ahem. That's a long-standing bug. The slope should be getting reset to 0 when you "Reset", so that it is 0 next time you calibrate. Thanks for reporting that. I'll fix it now.

Perhaps those slope values are only active with calibration values?

If a control is not being calibrated then it isn't even being looked at -- no changes whatsoever, just as if FSUIPC wasn't there. The slope is just part of the translation of "IN" values to "OUT" values that calibration performs.

Or they are reset when new calibration starts?

They should be reset if you've pressed the Reset (so the control is no longer even looked at) , and then go to re-enable FSUIPC calibration via the "SET" button. But there's evidently a bug which I shall fix. Sorry. BBut it won't affect non-calibrated controls.

my Force 3D Pro is a Force Feedback joystick to boot, something not supported by FSUIPC.

Ah, yes. That is a problem. Sorry about that. I have no idea how to support force-feedback.

But with no assignments made in FSUIPC, don't worry about using letters or numbers for joysticks in FSUIPC as it isn't relevant. If things are changing you need to see if FS is reassigning things sometimes, and consequently possibly altering your null zones and sensitivity settings. Else, I'm sorry, but I haven't a clue. Perhaps you could try without FSUIPC at all, because if it works fine without it I'd certainly want to know and would need to find out more.

Regards

Pete

Posted
Well, maximum dead zone -- it cuts out a huge portion of the axis capability.

Exactly, but this joystick seems a bit flaky especially when it is centered, more so than my previous joystick. As my hand merely rests on it, it is already slightly moving and for example sending an aircraft to the side when taxiing on a straight taxiway. To counter that, I increased the dead zone, so that I really have to move the joystick in order to get the aircraft moving. Same applies to the slew mode. I want to have to move the joystick significantly, before it moves the aircraft. To be honest, I do not fly by hand much, I have an auto pilot for that. So I'm most interested in getting accurate ground behavior with a firm-ish grip on the joystick. I don't want to be afraid to touch the thing in fear of looking like a drunk pilot.

:lol:

As to where I should make the necessary changes, that doesn't really matter to me, as long as it's working. So if I have to go to FSUIPC for that in favor of FS itself, so be it.

The GUID in the FS CFG extract you showed is this:

{9FC5FA50-0906-11DF-8001-444553540000}

See if that changes at all. That could certainly mess things about.

I will check and verify.

FSUIPC calibration has nothing whatsoever to do with FS's CFG file.

The confusion on my part remains. On one hand you say max sensitivity and min null, to get the most precise results. In other words, one influences the other. On the other hand you say, FSUIPC calibration has nothing to do with the FS9.cfg file. Sensitivity and Null Zone values are saved in the FS9.cfg file, as the various "AXIS_SCALE" and "AXIS_NULL" entries. I copied the section to illustrate my settings. If one has nothing to do with the other whatsoever, then surely, Sensitivity and Null Zone settings are completely irrelevant?!

:?

Erthe "Slope" button won't be there after a Reset because Reset cancels the calibration altogether. Do you mean after pressing "Set" again after "Reset", to re-enable the calibration? I'll have a look to see whether the Slope should be reset to 0 or not.

What I did was calibrate, set slopes. OK. Then back to FSUIPC, reset. OK. Then after closing FS, in the FSUIPC.ini file, the slope values (5 in my case) would still be there, I thought that was odd, so I mentioned it.

[LATER] Ahem. That's a long-standing bug. The slope should be getting reset to 0 when you "Reset", so that it is 0 next time you calibrate. Thanks for reporting that. I'll fix it now.

No problem.

Well, I'll go and check some things now. I'll be back with my findings.

Posted
The slope should be getting reset to 0 when you "Reset", so that it is 0 next time you calibrate. Thanks for reporting that. I'll fix it now.

Fixed in versions 3.967 and 4.583, which I'll upload to the Updates announcement later today.

Meanwhile, I've been experimenting along the lines you suggest here:

Now, when I close FS9 and then restart it without shutting down my computer, calibration seems to stick. If I close FS9, disconnect the power of the joystick, then reconnect and start FS, calibration seems to remain. But again, shut down my computer, start it up again, FS9, crazy joystick...

I've deliberately set the same sensitivity, dead zone as you, and assigned through FS9 only, not through FSUIPC. I also removed my "sensitivity mode" setting, to be sure it was all identical. I calibrated the slew fwd/back and left/right controls in FSUIPC and tested them. Yes, I agree, they are actually still usable despite the shenanigans FS has put the values through before they reach me. [More about why that might be in a minute ***].

The only thing I found wrong (and this is with FS9 only -- FSX seems okay), is that when re-loading FS and going into Slew mode, FS produces an erratic negative value on the elevator axis, which is of course the fwd/bwd axis. The value is way outside my calibrated centre "dead zone" (the FSUIPC variety), so the aircraft immediately dashes backwards at some huge rate. It is perfectly okay, though, as soon as i make any slight movement on the joystick.

I think that must be some residual value FS is using, and may be specific to my joystick and its driver (I use a MS SideWinder FreeStyle Pro for such testing).

Anyway, that was with simply reloading FS9 after closing it. No other changes. Then the same after Restarting Windows (using Win7 too, but 64 bit). Result: same as simply restarting FS9. So, another test: with the joystick disconnected during Windows' restart. Result Same.

Finally, disconnect joystick, power system off, Then back on, reconnect joystick when Windows is ready. Run FS9. Result: same.

So, sorry, but I reached no conclusions, except possibly that in your case wither Windows or FS9 is reconfiguring the joystick sometimes. I've no idea why or how except possibly it's that assigned GUID which is changing. Of course it could just possibly be something to do with that "Logitech Gaming Software (LGS)" you are also using. As part of the process of elimination, perhaps you could see if the same problems occur without that?

... One last thought on this. You do have the joystick connected and powered up BEFORE you start FS9, don't you? I think if you don't then it will be even more likely to treat it as a new connection and make its own choices.

*** On the point of how FSUIPC's calibration may be overcoming the problems I believe are caused by lower sensitivity and increased dead zone in FS, I think, in slew mode, you don't notice quite so much the loss of resolution you incur with those FS slider settings. FSUIPC's calibration still spreads the remaining values it does actually see across the whole range needed for full use of FS's slew controls, but with, possibly, only as few as 25% of the different values the joystick is capable of actually being used, you may simply be finding it easier to locate a fixed joystick position for a fixed-speed slew.

This is something I've not really experimented with before, because I find slewing infinitely (exaggeration perhaps ) more controllable using keyboard taps on the numpad. The only reason I provided separate calibration for slew was for those trying to slew from inside their cockpit, out of reach of a keyboard.

Regards

Pete

Posted

You last post crossed with mine ...

Exactly, but this joystick seems a bit flaky especially when it is centered, more so than my previous joystick. As my hand merely rests on it, it is already slightly moving and for example sending an aircraft to the side when taxiing on a straight taxiway. To counter that, I increased the dead zone, so that I really have to move the joystick in order to get the aircraft moving.

Yes, but if you are calibrating in FSUIPC, that's the whole purpose of the centre zone, the two values between which the output value is zero. By forcing that in FS you remove part of the original range of values available to you from the joystick. Likewise, an FS sensitivity of 50% merely causes FS to divide the value coming from the joystick by 2. Then FSUIPC calibration spreads the result back over the full range. In the end you are removing probably 75% of the stick's range of values and simply spreading the rest over the needed -16k to +16k range.

With many modern digital joysticks you might just get 500 or even 1000 values, if you are lucky. If they are based on ordinary potentiometers you'd be very lucky to get 200, and most only give about 40-100 values. Remove 75% of those and there might be only 10-20 different positions on the joystick which are effective.

Regards

Pete

Posted
You do have the joystick connected and powered up BEFORE you start FS9, don't you?

Yes, definitely.

Okay. Here's what I've found.

{9FC5FA50-0906-11DF-8001-444553540000}

That appears to stay the same after suqsequent boots and disconnections of the joystick. Good news I believe.

I changed Sensitivity to max and Null Zone to min. And added the stick sensitivity line to the cfg file (when FS wasn't running and I made no changes through FS's interface since). Those settings remain after boots and FS starts.

I calibrated both slew (in slew flight mode) and non-slew (in regular flight mode), added slopes according to my preferences. Although as I understand it better now, I will (also) have to change the center values (which are currently all +/- 512). But anyway, everything was fine as before, restarted the computer, etc... Back to the sim, slew mode, crazy joystick. I went into FSUIPC and looked at the IN and OUT values. Back/forward and left/right were changing normally as far as I could tell. However, the heading change wasn't. Even with a small twist of the joystick, both IN and OUT went straight to the max/min value. So either 0 or +/- 16000, nothing in between. And that was reflected by the crazy spinning around. Went out of slew, then back to FSUIPC. Most values behaved normally, except for the rudder, from 0 straight to the extreme. However, when twisting the joystick back to center position, the values suddenly came alive and behaved normally. Exit FSUIPC, back into slew, no more crazy spinning around...

I think that must be some residual value FS is using

I think that is happening to me as well. Although with another axis and not exactly the same, but still. The thing is, since I've been fixating on slew, I've always gone straight into slew mode. I'm now gonna test to see if the same thing happens, if I first move the joystick every way possible in normal flight mode. I have a feeling I will see no crazy slew spin after that, but I will check.

I may also try with and without the Logitech Gaming Software.

Posted
However, the heading change wasn't. Even with a small twist of the joystick, both IN and OUT went straight to the max/min value. So either 0 or +/- 16000, nothing in between. And that was reflected by the crazy spinning around. Went out of slew, then back to FSUIPC. Most values behaved normally, except for the rudder, from 0 straight to the extreme. However, when twisting the joystick back to center position, the values suddenly came alive and behaved normally. Exit FSUIPC, back into slew, no more crazy spinning around...

The symptom of only having the two values, full on or full off, or full up and full down, or whatever, is usually one created by an axis operating in "digital" mode (on/off only) rather than analogue mode. Many toe brake setups are like this, for instance, and I think it is configurable on several makes of joystick and pedals.

I haven't yet tried the heading slew here on FS9 (no normal joystick pedals), but I could try assigning the little throttle wheel to it I suppose.

One other thought. If powering the PC off and on does it, but doesn't change the GUID and therefore FS's assignments and settings, the only other thing it could be down to is surely the power off and on of the joystick and/or its USB connection?

So, does the same problem occur if, instead of powering the PC down and back, you simply power down your joystick and disconnect its USB connection, then connect it all back up again? If so it would be some sort of initialisation bug in the device or its driver. If it only occurs with the PC power cycling then there must be something odd happening in either the driver or the Logitech software which only occurs once.

[LATER]

Tested heading slew, and that's okay. But it isn't the FS control that has the problem, at least in my case. It is the axis input from the joystick. Only one of the two axes sends a spurious initial value. I logged the Axis controls (an FSUIPC logging facility -- see Logging tab) and got this:

31015 *** AXIS: Cntrl= 65869 (0x0001014d), Param= -2938 (0xfffff486) AXIS_SLEW_HEADING_SET

34312 *** AXIS: Cntrl= 65869 (0x0001014d), Param= 0 (0x00000000) AXIS_SLEW_HEADING_SET

That's with my Elevator axis assigned to the heading slew. It settles to zero immediately I touch it. And if I assign the same axis to the Forward/Backward slew, the exact same incorrect value arrives for that.

However, if I don't calibrate via FSUIPC, or remove FSUIPC altogether, that spurious value doesn't arrive. Which is puzzling me, so I am now going to delve into that. Perhaps you could try the same and let me know if your problem is the same thing.

Regards

Pete

Posted

FYI, this is what the thing looks like: http://www.logitech.com/index.cfm/gamin7&cl=us,en

Without the Logitech Gaming Software active (though still installed), sensitivity is somewhat different, but the same problem occurs.

With or without LGS, I've now been able to consistently avoid the issue by first moving the joystick thoroughly. After that, going into slew, everything behaves normally. As for what causes it precisely, I have no clue and no desire to find out to be honest. I'm already glad I know what to do to avoid it, after all this testing.

I'll finetune the calibrations a bit and then I'm gonna do something else, anything else.

:lol:

Thanks for your patience and help.

:)

Edit: simultaneous post/edit.

Which is puzzling me, so I am now going to delve into that. Perhaps you could try the same and let me know if your problem is the same thing.

As per above, maybe tomorrow.

:wink:

Posted
However, if I don't calibrate via FSUIPC, or remove FSUIPC altogether, that spurious value doesn't arrive. Which is puzzling me, so I am now going to delve into that. Perhaps you could try the same and let me know if your problem is the same thing.

Okay. What is happening here is that, because I calibrated the slew axis but not the equivalent elevator axis, and my null "centre" zone is actually well into the +ve side of the axis values (the device sits on the desk in a pitch up angle), a value FS has for the elevator is being seen initially when changing to Slew mode, and this is then calibrated into a -ve value, so it slews backwards.

It does this every time I change into Slew mode with the device sitting on the desk. It is a direct result of the same axis being used for two different controls with two effectively different calibrations, and can be fixed by calibrating both the same. The reason that FS itself, without FSUIPC, doesn't have the same problem is exactly because it doesn't calibrate, it just uses the Windows calibrated values from the driver which are naturally the same no matter which FS control is assigned.

I would certainly conclude that what I am seeing is not at all related to what you are seeing. So, sorry, I have no solution. It seems to be something outside of both FS and FSUIPC control.

Regards

Pete

Posted

Mike

You said in your first post: "I have a Logitech Force 3D Pro since a couple of weeks, it needs power in addition to a USB connection. I do not always have its power on and so the joystick is not "connected" to my computer all the time, I think."

I'm not sure if Pete addressed this but every time that FS9 starts and the joystick is not connected and then re-connected it is seen as a "change" and the SIM resets all the values for that controller and it may assign spurious duplicate assignments.

I use a G940 (needs a power supply) which uses similar Logitech software, so I do 2 things and I keep it powered on at all times.

1) Any assignments in the Logitech software I make sure that they are exactly the same in FS9 and I clear any duplicate assignments.

2) Any assignments that I make in FSUIPC - I make sure that those assignments are not also made in FSX and/or the Logitech software.

My G940 works perfectly in FS9/FSX using the logitech software with FSX and FSUIPC/FSUIPC4, selectively as described.

Pete is correct for the joystick to work optimally you must have ALL axes (some are hidden in the drop down box) Sensitivity - full right - (100%) and Null Zone full left (0%). These can be changed but I wouldn't advise it - the realism settings in the SIM work much better in this configuration. When you are setting up the joystick make sure that your realsim settings in FS9 are all full left ie "Simple", you can then increase the complexity as you get more proficient.

This probably won't help but it does give you another POV.

Regards

PeterH

Posted
You said in your first post: "I have a Logitech Force 3D Pro since a couple of weeks, it needs power in addition to a USB connection. I do not always have its power on and so the joystick is not "connected" to my computer all the time, I think."

I'm not sure if Pete addressed this but every time that FS9 starts and the joystick is not connected and then re-connected it is seen as a "change" and the SIM resets all the values for that controller and it may assign spurious duplicate assignments.

Let me put it differently, the usb cable of the joystick is always plugged in. The power isn't always turned on. So I was wondering if without that power, the computer still sees the device? I'm guessing not. Which may have given identification issues. My previous joystick was always connected (to same port) and didn't need power, so as far as the computer was concerned, it was always there, always the same, no likely id problem.

I do always turn on the power before starting the sim.

As far as this not being the same as Pete's issue, he's probably right. The problem I have is not unique to this joystick however. I didn't put two and two together until last night, but my previous joystick was the MS Sidewinder, it worked with the sidewinder software (4.0) or whatever it was called. It included a "calibration" utility and every time I opened it, the throttle would indicate 100%, so full throttle, despite the lever's position. It wasn't until moving the lever, that the correct value was displayed.

Oh well.

:wink:

Posted
... every time I opened it, the throttle would indicate 100%, so full throttle, despite the lever's position. It wasn't until moving the lever, that the correct value was displayed.

Actually that's a common issue with many USB devices, because to save on traffic they only send position data when it changes. It was different with the old Game Port devices because they had to be explicitly polled.

It was such a common problem that in one update last year I dealt with it in FSUIPC assignments by deliberately discarding the first few readings I got from an axis, both at the startup and after changing anything or reloading settings in the FSUIPC options, and then waiting for a changed value. Unfortunately I cannot do that for axes assigned outside FSUIPC because it isn't my code reading them. And in any case all that means is nothing happens on an axis until you move it, so it isn't quite so different as what you now have to do, except it may avoid things zooming off in Slew mode if your FS is initialising into that mode.

I didn't associate your problem with that common problem because you said it was only wonky after powering the PC down and up again. The usual problem occurs every time a fresh connection is made to a USB HID device. But maybe the drivers for yours are different. Maybe they keep a memory of things as long as power is available.

Regards

Pete

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. Guidelines Privacy Policy We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.