ForceFeedback in DCS - realistic or gimmick?

I've been asking this years ago and never got an answer. Let me ask the question once more as more and more direct control link A/C are being released. By realistic I mean proportional to hinge momentums on specific control surfaces as calculated by the flight model.

Probably more realistic than with any non-ffb stick...

I tried FFB once (G040) and didn't like it, I ended up returning the gear to Logitech.

If FFB could be done well by the controls it would be a great idea but it's implementation on game gear is that it feels like exactly what it is, pushing the stick against cheap electric motors. It feels like stirring a bucket of rocks with a stick. Also there's a significant dead zone in the forces at the center. It makes aiming accurately just about impossible although maybe that's realistic. It doesn't feel real though, it felt like using a broken stick. Maybe it was in fact defective.

With Microsoft Sidewinder FFB2 FFB is really great in DCS. You can feel control forces at high speeds etc. Of course FFB sticks and wheels will never be fully realistic but they will compensate some missing things you would feel in real life. I would never ever play any sims without FFB anymore...

I tried FFB once (G040) and didn't like it, I ended up returning the gear to Logitech.

If FFB could be done well by the controls it would be a great idea but it's implementation on game gear is that it feels like exactly what it is, pushing the stick against cheap electric motors. It feels like stirring a bucket of rocks with a stick. Also there's a significant dead zone in the forces at the center. It makes aiming accurately just about impossible although maybe that's realistic. It doesn't feel real though, it felt like using a broken stick. Maybe it was in fact defective.

Not broken. Thats how g940's are. You can reduce the notchiness but not the dead zone.

MSFFB2 is nicer feeling.

The old Logitech wingman force probably had the best FFB ever, really strong motors and the drives were done with a cable loop (no gears).....smooth as silk, I have the carcass of my old one waiting to be spliced with my G940 in some way.

I appreciate the responses but guys, I'm not asking how are your impressions with your FFB sticks in DCS. I'm asking whether anyone has noticed some FFB effects that would indicate DCS's FFB implementation is to 'add hollywood' to the flying rather than mirror real effects. An example would be: stick shaking at the top of a hammer-head stall turn maneuver. Or: shaking of the same frequence regardless of whether the stall is at low speed or at high speed (accelerated stall).

I appreciate the responses but guys, I'm not asking how are your impressions with your FFB sticks in DCS. I'm asking whether anyone has noticed some FFB effects that would indicate DCS's FFB implementation is to 'add hollywood' to the flying rather than mirror real effects. An example would be: stick shaking at the top of a hammer-head stall turn maneuver. Or: shaking of the same frequence regardless of whether the stall is at low speed or at high speed (accelerated stall).

FFB are not just hollywood-like effects in DCS. They are trying to replicate the physically correct sensation of what a pilot would experience, as far as I can tell.

But to be honest, FFB seems to be still quite "WIP". Every single patch of the last 1-2 years fixed some issues and/or changed some behaviour of how FFB is handled in DCS. It seems to me that the results also depend not insignificantly on what FFB stick you use (i.e. G940, FFB2). In my eyes there is still some work to do on EDs side, but the FFB integration is in fact getting better and better, as far as I can tell.

Never noticed anything other than control loadings and buffet (which is probably exagerated)

Forces are FAR too weak though.

Never noticed anything other than control loadings and buffet (which is probably exagerated)

 

Forces are FAR too weak though.

Those will probably never be on a realistic level. I mean, hey, our joystics are made of PLASTIC ...

Never noticed anything other than control loadings and buffet (which is probably exagerated)

 

Forces are FAR too weak though.

I think that's a limitation of the hardware though.

OK, thanks.

Do you know of any way to analyze what signals a game actually sends to FFB stick? In DCS or another way.

Those will probably never be on a realistic level. I mean, hey, our joystics are made of PLASTIC ...

mines not...but I agree its like stirring a bucket of rocks:music_whistling:

mines not...but I agree its like stirring a bucket of rocks:music_whistling:

Well, ok, you got me ... what IS your stick made of and where did you get it (and how much did it cost?)?

Hard to say. The only plane I've flown with a stick instead of a yoke had negligible stick forces. Also, there was no stick shake at all before a stall.

What does force feedback feel like in a real jet that has it? Like the F-16? I assume it's not called Force Feedback and instead it's called something else.....

What does force feedback feel like in a real jet that has it? Like the F-16? I assume it's not called Force Feedback and instead it's called something else.....

F-16 has force sensing which is something completely different. Basically the stick registers how much pressure you apply to the stick and the computer moves the control surfaces based on that. The stick doesn't even have to move for that although I believe it does move a little bit in the F-16.

F-16 has force sensing which is something completely different. Basically the stick registers how much pressure you apply to the stick and the computer moves the control surfaces based on that. The stick doesn't even have to move for that although I believe it does move a little bit in the F-16.

Oh cool thanks

One of the main reasons I prefer FFB sticks is because of the realistic trimming behavior, but the DCS warbirds have pretty good force effects. However, I think there's some room for improvement in all the aircraft. There is an excellent FFB program for FSX called FSForce. It allows control and tuning over the speed/control force curves, friction & damping adjustments, along with environmental effects like pavement bumps. I think something like that for DCS would be excellent.

I have a Saitek X52 handle on the mechanics of the MS FFB2 which works pretty well, I also hacked a saitek cessna yoke with control forces provided by MS FFB2 guts, which has been working quite nicely in FSX.

I won't ever go back to a spring stick, but I wish there were some good modern commercial products available. Until then, we'll just have to keep hacking existing hardware.

OK, thanks.

 

Do you know of any way to analyze what signals a game actually sends to FFB stick? In DCS or another way.

Oscilliscope would be a start I guess

Oscilliscope would be a start I guess

I think, starting at OSI layer 6 or 7 instead of 1 would probably save some time. :o)

It would still be interesting to have somesort of access to the force feedback information that DCS sends to the stick. I don't have alot of free time right now, but I've been playing with the idea of hooking up an Arduino (Leonardo version that can act as an USB Input device) and have it registered as axis input device in DCS. As far as I understand it, FFB commands are midi protocol commands (The stuff that's used for the old gameport keyboards and music boxes like the MT32) send to the device. Normal sticks ignore that stuff. If you can hook up an Arduino to read that, you could fashion some really nifty stuff like a directional buttkicker or even have a pully setup for your stick in a pit.

It would still be interesting to have somesort of access to the force feedback information that DCS sends to the stick. I don't have alot of free time right now, but I've been playing with the idea of hooking up an Arduino (Leonardo version that can act as an USB Input device) and have it registered as axis input device in DCS. As far as I understand it, FFB commands are midi protocol commands (The stuff that's used for the old gameport keyboards and music boxes like the MT32) send to the device. Normal sticks ignore that stuff. If you can hook up an Arduino to read that, you could fashion some really nifty stuff like a directional buttkicker or even have a pully setup for your stick in a pit.

Huh? Todays joysticks, FFB devices included, are USB devices. You access the devices capabilities through means of DirectX which in turn talks to the device driver ... which then, via USB cable, talks to the firmware of the stick.

To analyze how exactly what FFB effects are controlled, you would need a (virtual) device driver that allows you to monitor the API calls. Maybe there already exists something like that? Or perhaps in form of some DirectX debugging tools? Maybe the SimFFB sourcecode would be a good start (but I not sure if theoretically SimFFB would be able to receive FFB commands, though)

Oscilliscope would be a start I guess

You'd need a FFB stick for that but good idea anyway

I think, starting at OSI layer 6 or 7 instead of 1 would probably save some time. :o)

I have to duck it. Never heard of it before.

It would still be interesting to have somesort of access to the force feedback information that DCS sends to the stick. I don't have alot of free time right now, but I've been playing with the idea of hooking up an Arduino (Leonardo version that can act as an USB Input device) and have it registered as axis input device in DCS. As far as I understand it, FFB commands are midi protocol commands (The stuff that's used for the old gameport keyboards and music boxes like the MT32) send to the device. Normal sticks ignore that stuff. If you can hook up an Arduino to read that, you could fashion some really nifty stuff like a directional buttkicker or even have a pully setup for your stick in a pit.

Good ideas! Except the interface has to bee USB+DirectX, as Flagrum said.

Or perhaps in form of some DirectX debugging tools?

More good ideas.

I'd really like to hear from the developers on this though. Why waste countless of hours investigating this when a developer can answer this in 10 minutes? SithSpawn, are you there? ;)

Huh? Todays joysticks, FFB devices included, are USB devices. You access the devices capabilities through means of DirectX which in turn talks to the device driver ... which then, via USB cable, talks to the firmware of the stick.

 

To analyze how exactly what FFB effects are controlled, you would need a (virtual) device driver that allows you to monitor the API calls. Maybe there already exists something like that? Or perhaps in form of some DirectX debugging tools? Maybe the SimFFB sourcecode would be a good start (but I not sure if theoretically SimFFB would be able to receive FFB commands, though)

Sry if I was a bit unclear, let me precise what I meant (and as with everything on forums, if I'm wrong, please correct):

Yes, todays sticks are USB. The FFB protocol (Aka, the way FFB commands are transmitted) is still the old midi protocol from what I was able to read about it. That's due to the history of the technology which was introduced on the old midi/gameport 15-pin ports which back then were used for sticks and midi devices. Like on this beast which I still have around here somewhere:

What they did was use the midi capabilities of the port, which the sticks didn't use anyways, to transmit the control commands for the FFB motors over it.

I'm assuming right now that with todays FFB USB sticks, that didn't change. It's still midi commands that are being sent to the device.

(Addendum: DirectX is just a standardized layer that generates those midi commands so you don't have to directly tell the device how to run its motors. It's not replacing the actual protocol from what I understand about it. Corrections of course welcome if I got it wrong.)

About the software device: Yes, you could of course do that. But what you would really want is make use of the information in a haptical way (aka, you want to feel something, not just display a number on your screen somewhere). Hence the idea to put an Arduino device behind it which you can use to put the FFB information into effect.

A software device to log the commands would be a good first step though. I'm not privy to Windows Device Driver Developement, maybe somebody else can comment on the feasibility of it.