+- Forums (https://www.vdrift.net/Forum)
+-- Forum: Community (https://www.vdrift.net/Forum/forumdisplay.php?fid=3)
+--- Forum: General Discussion (https://www.vdrift.net/Forum/forumdisplay.php?fid=8)
+--- Thread: Hardware: steering wheel and pedals? (/showthread.php?tid=403)
- stenyak - 02-22-2007
There are several "momo force" wheels (i think they were 2?). You prolly got the other one
- eckzow - 02-23-2007
Not to piggyback or anything, but speaking of the G25, has anyone figured out the ~300ish degree restriction? The wheel turns itself the full 900 degrees when it first plugs in, but in the (admittedly small amount of) time that I've played with the wheel it doesn't seem to respond to more than 300 degrees (nor do I get data from the clutch, nor any gear except 1st and 2nd)...
- stenyak - 02-23-2007
I guess you mean only in linux?
- Nigo - 02-23-2007
stenyak Wrote:There are several "momo force" wheels (i think they were 2?). You prolly got the other oneno, just one. The Momo Force is the red one, the Momo Racing FFB is the black one
the first was better (and more expensive)
- thelusiv - 02-23-2007
How was it better? I have the black one, and looking at pictures of the red one, it looks like it doesn't have the (useless) "shifter" knob, and the things to attach it to a table look like they probably don't work as well as the ones on mine. I can't say anything about the hardware though since I've never used one, but they look almost identical in terms of casing, the wheel itself, the buttons, etc.
eckzow, hi and welcome.
I haven't got a G25 so I'm not sure what the problems with it are in Linux, or how exactly to solve them. I'm guessing that since the previously available Logitech wheels had less features, that the new one is limited by the driver capabilities. It sounds like someone needs to see just how it works in Windows, probably spy on the USB traffic some, and use this data to improve the Linux driver. If I had one of these wheels I'd be doing just that...but I can't afford one. 
So I hope someone does this before they become cheap, because I will certainly get one when that happens...
- stenyak - 02-23-2007
The expensive one had better materials, and i think even an optical pot in the wheel. I think the red one was the good one.
The problem with G25 is exactly that: it has more features:
-900º of steering, instead of the usual 200+º.
-Heaps of buttons, including an 8-way hat.
-An H-shifter with 7 regular gear positions (6+1) + neutral (=="no gear").
-A switch so that the H-shifter physically behaves like a rally car sequential shifter (only up-down movement).
- eckzow - 02-24-2007
Whew, active community. That's refreshing... anyway, sorry I haven't been keeping up as well, but I was at work all day and I forgot to change my password to something I knew last night so I couldn't log in until I got home..
Let's take these one at a time.
stenyak Wrote:I guess you mean only in linux?
Yes, I meant the default 200 degree set points that the wheel initializes to when it is enumerated. After getting the basic force feedback utils working I can confirm that they work on the wheel (ffcfstress, etc.). I didn't notice any particular problems with them, not even the "force feedback stopped working after a while" problem I've seen posted in a few other places.
thelusiv Wrote:eckzow, hi and welcome.
Hi, and thanks! I happen to have the good fortune of access to a G25 wheel for the foreseeable future and probably a few good hours a day to play with it. I've finally gotten out of the "find someone to do it for me mode" so I'll start the reverse engineering process to find out how the bad boy works.
stenyak Wrote:The problem with G25 is exactly that: it has more features:
-900º of steering, instead of the usual 200+º.
-Heaps of buttons, including an 8-way hat.
-An H-shifter with 7 regular gear positions (6+1) + neutral (=="no gear").
-A switch so that the H-shifter physically behaves like a rally car sequential shifter (only up-down movement).
Yeah, I guess I was hoping that the USB-HID protocol was abstract enough that most of those things (like the extra axes, etc.) would be handled in stride. Naive, eh?
So far things aren't too bad though: something in the fftest demos changes the steering calibration, so that's a starting point. Most of the heaps of buttons are handled by standard HID stuff, the exceptions being the shifter (first and second work, but the others do not).
The H-shifter to sequential shifter switch doesn't seem that complicated. It seems to use the same "buttons" as the H shifter does for first and second. The one trick might be detecting which mode it is in, though.
The hat shows up as a set of axes, just like the d-pad on the logitech rumble joypad.
Most depressing is the clutch, which doesn't seem to show up anywhere I've seen so far.
All of the controls though, including a slew of force feedback effects (like wooden bridge, etc.) as well as things like configuring the wheel's set points (apparently anywhere from 40 to 900 degrees) are all available through the game controller configuration in Windows. I like this, because it means I'm not going to have to jump through hoops to get good I/O on this thing.
Anyway, I think this project is within my abilities, so I'll certainly give it a shot. I plan on familiarizing myself with the current drivers for the wheel this weekend and come the beginning start trying to reverse some protocol.
Wish me luck!
- stenyak - 02-24-2007
Wohoo! Cool, thanks for your work! I actually own a g25, so i can test it whever i want; if you need some testing just let me know, and supposing i have enugh time, i'll be happy to help
Edit: i think the H-to-sequential shifting switch cannot be detected via software, none of the sims i've tried can do it.
- Nigo - 02-24-2007
thelusiv Wrote:How was it better?it had steel bearings, 2 potentiometers, leather, and the pedals were a bit better
- eckzow - 02-24-2007
stenyak Wrote:Edit: i think the H-to-sequential shifting switch cannot be detected via software, none of the sims i've tried can do it.
I agree that it might not be present because you could just remap your controls accordingly, but they dedicate a whole LED on the unit to displaying your shift-mode status, so you'd think they'd dedicate a bit of their output format to it too... *shrug*
- thelusiv - 02-25-2007
Nigo Wrote:it had steel bearings, 2 potentiometers, leather, and the pedals were a bit betterI see, that does indeed sound better. It doesn't look better though. All that hideous red...
Oh well, hopefully I'll find a cheap Driving Force Pro soon, and then...900 degrees of goodness! 
- eckzow - 03-12-2007
Sorry I haven't been updating you guys on my endeavors with the wheel, but I've got a lot going on that's keeping me away from working on things.
I haven't quite figured out the clutch/shifter yet but I have managed to unlock the full range of the wheel (it can be configured to run from 40 to 900 degrees). Eventually I'll get around to learning the hiddev stuff well enough to integrate that into the driver somehow...
This isn't really the appropriate place for this, but since there seemed to be some technically skilled (enough to recompile a kernel anyway) people here... if anyone cares, unlocking rotation is simple. All you need to do is send 7 bytes to endpoint 1 on the device:
0xf8, 0x81, aa, bb, 0x00, 0x00, 0x00
Where aa, bb is a little-endian unsigned short valued somewhere between 40 and 900. For instance:
char deg40_msg[7] = {0xf8, 0x81, 0x28, 0x00, 0x00, 0x00, 0x00};
char deg900_msg[7] = {0xf8, 0x81, 0x84, 0x03, 0x00, 0x00, 0x00};
Would be the messages to set the soft-stops to 40 and 900 degrees respectively. Currently to do all this I manually unbind the usbhid driver, claim the interface, send the message, and then rebind the usbhid driver.
If anyone has questions or wants help, just post away.
- thelusiv - 03-12-2007
That's great news! eckzow, if you haven't already you should post this info to the linux-input mailing list. There are links and directions for joining the list in posts either in this thread or the force feedback thread.
- stenyak - 03-13-2007
Cool news
Thanks for your work!
G25 FF protocol info - anrp - 05-10-2007
I meant to post this some time ago, it's what information I've gleaned about the G25 FF protocol. .. I can answer some questions about it I guess, I tried posting to the linux-input list but I couldn't figure out how to subscribe (surely I'm missing something...)
edit: noteworthy is the fact that most messages are supposed to be 8 bytes long, the 7 byte long ones are an artifact of how I got them.
Code:
starts up in driving force mode
f810 - write to plugin device (046d:c294) - causes device to disconnect and reconnect with a new ID
(046d:c299) this is necessary to get at the clutch axis and the additional shifter buttons
f801 - disconnects and reconnects self; changes device to 046d:c298 - driving force pro mode
f8813804 - set turning angle full scale
____^^__ - low
______^^ - high (is degrees, this one is 900)
fe0b0101ff - centering spring, slow spring ramp
____^^____ - left ramp speed
______^^__ - right ramp speed
________^^ - overall strength
axis wall effect
210b8080 f010ff - right side maximum suck
210b8080 f000ff - right side maximum push
210b8080 8000ff - right side half push
210b8080 8010ff - right side half suck
210b8080 0801ff - left side half suck
210b8080 0f01ff - left side max suck
210b8080 0f00ff - left side max push
210b8080 0700ff - left side half push
210b4545 0f88ff - offset -0.4563, left side, max push
210b4545 0f89ff - offset -0.4563, left side, max suck
210b0000 0f01ff - offset -1, left side max suck
210bc1c1 0fcdff - offset 0.5145, left side max suck
210bc1c1 0fccff - offset 0.5145, left side max push
210bc1c1 f0ccff - offset 0.5145, right side max push
210bffff f0eeff - offset 1, right side max push
first hex digit is effect num (1-f)
second: 1 -> start, 3-> stop
5 -> cs effect stop, 4 -> cs effect start (?)
effect 1 is the force motor, effect 2,3 seem to be a one-side force motor (?)
byte 2: no idea, mostly 0b, sometimes 00 when stopping
1-f - effects
--
force effect
11088180 00
__^^____ __ - always the same (effect type..?)
________ ^^ - when set to 0 or 80, no effect. when set to 7f, max left, 81, max right
____^^__ __ - force strength... 81 is zero pos, ff max one dir, 01 max other, 00 = ?
> 81 is turn left
----
spring effect -
210bdeff 26e0ffff
____^^__ ___?____ - offset from left, left side spring (low 3 bits at ?)
______^^ __?_____ - offset from left, right side spring (low 3 bits at ?)
________ ^_______ - right side spring strength ramp rate
________ _^______ - left side spring strength ramp rate
________ __^_____ - (bit 1) if 1, pull instead of push
________ ___^____ - (bit 1) same
-----
slope effect
210b8080 2200ffff
________ ____^^__ maximum spring force for right
________ ______^^ maxmimum spring for left