Microsoft Sidewinder Force Feedback 2 - deadzone?

[LuCiPh33R]

Hi folks, just in the process of getting my setup the way I want it. I very much like using my Nexus 4 as a poor man's EDTracker!

I've inherited a Microsoft Sidewinder Force Feedback 2 (last used for several playthroughs of Freespace 1/2 - good times), but now that I'm paying more attention to these things, I'm noticing a small but still quite significant deadzone in the stick in the pitch axis (up/down). These things use optical sensors, so I figured that in theory there should be no deadzone at all. However, I did manage to find someone suggesting that this is firmware-encoded (sorry, can't post links yet).

I'm running Windows 8.1x64, so the MS Sidewinder utility software obviously doesn't work anymore, and you only get access to basic settings in Windows. simFFB doesn't seem to offer much in the way of control there. It seems that some people claim there is zero deadzone in *their* MSSWFFB2, so I'm now starting to wonder if this is an age/batch thing - mine is the "old" grey-triggered model, not the newer one with a red trigger, so I wonder if that has anything to do with it. The roll axis (left/right) has a deadzone small enough that it doesn't matter, but rather curiously nose down (forward) has a larger deadzone than nose up (back). The deadzone almost coincides with the region where obvious force-feedback auto-centering does not function, but not quite. I suspect part of the asymmetry is due to the way the joystick is centered without the force feedback. If you compare this to the power cord unplugged though, the motors are definitely doing something in the middle, but it is definitely very subtle. I wonder if its possible to tweak something internally to adjust this, but my google-fu fails me.

All the deadzones in ED are set to 0. Its not huge, but it is making me consistently overcompensate when trying to make small corrections in dogfights. I haven't touched any of the joystick settings in ED yet - any suggestions on the best place to start to try to work around this issue? Does the T16000m have less of a deadzone? Do I even want zero deadzone?

Thanks!

 

[LuCiPh33R]

Since no-one has responded I thought I'd follow up...


Previous thread on the SWFFB2 in general: https://forums.frontier.co.uk/showthread.php?t=21608
Speculation about hard-coded deadzones: http://forums.eagle.ru/showpost.php?p=2234108&postcount=58
Interesting technical info: http://riseofflight.com/forum/viewtopic.php?f=350&t=29129


It has been suggested elsewhere (simFFB threads?) that without the software, everything defaults to their minimum/maximum values. Screenshots of the old software (http://www.activewin.com/reviews/hardware/joysticks/microsoft/ff2/advfeatures.shtml) definitely show the ability to control the "inertial zone", but its not clear if we also get access to the deadzone settings too. Hopefully (or not?) we get the minimum ones in the absence of the driver. This would mean getting the Sidewinder software working would make no difference at all.


Pictures of other people dismantling it:
http://www.simprojects.nl/ms_siderwinder_ff2_hack.htm
http://bbs.hitechcreations.com/smf/index.php?topic=297141.15 - modified philips screws - use a good #2!


I tried to dismantling myself, and can confirm that the sensors do indeed appear to be pots, albeit apparently very good ones. In the process of dismantling it, I think I managed to get something stuck in one of the gears, which then started binding. This seems to have been fixed by me scraping out the teeth with some tweezers, and applying some fresh grease. And some new pants!


So now I'm none the wiser as to what's wrong.
1. It's a driver and/or firmware issue. Could be investigated by installing XP and locating the drivers from somewhere, which so far I haven't managed to locate. I'd have to start messing about with dual booting and disabling UEFI secure boot etc. etc. That the "inertial zone" and the extra bit of deadzone coincide suggests that this might be it, in which case there doesn't seem to be much I can do about it.
2. It's a dying pot. No-one seems to have found a source for these, so I'd have to buy a suitable donor stick. £££! Adapting other ones (e.g. http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=103_81&products_id=193) might be tricky since the SW pots have a "gibbous moon" shape to the ends.


Without having another working stick to compare to, it's hard know whether I'm just being super picky or it really is something not working right! I am reluctant to start resoldering in case I break an otherwise perfectly functioning joystick.

Hopefully someone will find the info useful...

 

[LuCiPh33R]

Once again replying to myself!

I've been playing around with simFFB (http://forums.eagle.ru/showpost.php?p=1628706&postcount=43), and closest setting to the defaults I could get was 92-95% spring force (definitely not 100%) and 0% damping and friction. I couldn't get myself to like the damping or friction effects at all, but I think there's definitely some value in dropping the spring force to 65-70% - the stick is less fatiguing that way. Any less and the auto-centring doesn't work very well IMHO.

DIView (http://www.leobodnar.com/shop/index.php?main_page=page&id=12) is a rather nice utility to see what the joystick is doing, and the y deadzone issue is more obvious that way, as well as the deadzone at the ends of the throttle.

Universal Joystick Remapper (https://github.com/evilC/AHK-Universal-Joystick-Remapper) installed OK, but the required vJoy (it doesn't see the version of vJoy that opentrack uses) installation hung on the last part (W8.1x64), and UJR claimed I was using an old version, so I couldn't play around with that. I couldn't be bothered to uninstall opentrack to find it if it was fighting with the previous installation.

 

[JamesNova]

Sorry to necro this thread, but I like your devotion to this topic! I have one of these sticks myself I picked up last year, almost unused! I used to own one but I asked a "friend" to return it to the computer store he worked at because it was dysfunctional but he ended trading it off for drugs. Can you believe it... If only he knew what sacrilege he was committing.
As far as I was aware, these sticks are optical. I'm not sure about the qualities you're experiencing compared to my own, but I play DCS World with it, and don't notice any particular deadzone. The motors aren't 100% accurate, but they're close enough. In DCS I also fly helicopters (the effort required to keep them in the air must be finely tuned and its incredibly laborious - as I imagine they would be in real life) and thought any deadzone issue would be most notable in that situation. I would suspect software though, as I can't imagine a reason the hardware would have a deadzone in the first place (that would be very shoddy design). In DCS, you can create a deadzone of your choosing, but I set it to 0 for the helis as they don't really have a dedicated center point with force feedback. I'll have a look for some more info and get back to you if you're still interested. I did find some interesting articles detailing the insides of the ffb2 as some guy decided to stick two of the drive mechanisms together to create a more powerful, extended version of the stick which he was rather proud of despite how hideous it was! I'm amazed he got it right but it turned out to be much more than he bargained for.

Did you ever get back to finding out what was going on with your stick?

Also I wouldn't set any friction or damping for Elite Dangerous. It's great for flying a helicopter but not much else. Also the friction effect feels very lumpy and is not suited to fine controlled input scenarios such as flying anything with realistic physics. I'm happy you shared those links, thanks for those - I'd like to see what I can do with this stick because DCS doesn't really support it and neither does practically anything recent as far as I know. What a shame - it was a fantastic FFB stick if there ever was one. I'm very protective of the one I have today.