GPL FF settings for MSFF?

I was playing around with the FF settings in GPL last night, and found they have a strong influence on my lap times, as much as 3 seconds at Rouen. For the past month or so, I've been running what felt like pretty realistic settings, but that really kills my hands and wrist in long sessions. Lighter settings were OK for a short while. I no longer needed a death grip on the wheel to fight the strong kicks, and felt I had better control for fine corrections. The road forces, though, seemed to be strongly muted, and I could no longer feel what the front was doing. What started as a short experiment finally ended in frustration many hours later in the wee hours this morning. I'm no closer to understanding the settings in core.ini, and now don't recall what I had them set to. What do you use for your MSFF? Michael.

GTX_SlotCar <[email hidden]> wrote in message news:Ctvu4.292$[email hidden]... > What settings are you using now and what did you use before? I'm using a latency of 0.010, and a damping force of 35.0. Max torque is commented out. I'm looking for best feel for road conditions, without inviting repetitive stress injury. Before, the forces were high enough that my hands and wrist were still sore the next morning. Two weekends ago, I ran a full GP at Spa plus the second 2 hr practice before that, about 3 hours of actual track time in one sitting. (I finished 5th, and was pleasantly pleased, if tired and drained. I had only recently made some strong breakthroughs in lap times.) I think damping force was 85, and possibly the initial default 0.035 latency; max torque was still commented out at the time, I believe. Part of the problem is I didn't make note of how it was set before I started making changes (who would think you would so easily forget 3 little numbers?). The wheel was heavy, and kicked strongly from time to time; road feel was very good, and the shaker let me know when I was running right out to the pavement edge. Anyway, I've since tried changing the FF numbers to see if I could get a lighter feel without losing too much feedback. I never did find a good value to use for max torque. It would either shimmy strongly, or was too light to be of much use. By shimmy, I mean the wheel would slap back and forth if I let go, even at slow speed. At one point, I could distinctly smell the ozone of an overstressed electric motor. I realize that the shimmy is an artifact; not finding resistance, the aligning forces would quickly turn the wheel until it reversed, and then it would feedback the other way. However, I never did find a balance that I liked. So, that's why I'm asking. Last weekend, I thought I would amuse myself by taking a break from struggling around Rouen, and take a few turns around Spa while loaded up on a half bottle of Merlot (Lynfred Winery 1999 Private Reserve; absolute nectar if you like this kind of thing!). My second lap out, I turned a PB of 3:21, 0.8 seconds faster than ever before. It was at this point that I started to monkey with the FF numbers, and things haven't been right since. Even if I can't improve on it, I'd like to at least get things back to where they were. Hence my appeal for your help. My lap times, at both Rouen and Spa, are now almost 2 full seconds slower than at that peak of alcoholic bliss. The only changes were to the FF settings. Michael.

What settings are you using now and what did you use before? Slot

Lighter This is normal. When I first started racing karts, my lower arms would absolutely kill after just 15 mins of racing. Just imagine hauling around an F1 car with 1 turn lock to lock and no power steering. I still use the original ff settings because I don't really understand what affects changing those numbers will have, and I don't have the time to try all the different combos. If anyone could suggest a 'realistic' set of numbers to use, that would be great.

By golly and gosh! these seem to have done it. Thanks for your help. It's very light, but tremendously driveable. Great feedback without undue strain, and it just takes two fingers to guide it down the long straights. Oh, and the sharp difference between entrance at Masta and mid-point is frightening! I don't know how I managed at all before this. Michael. GTX_SlotCar <[email hidden]> wrote in message news:b%Zu4.5$%[email hidden]... > Try these settings, what have you got to lose? > Damping=4000 > Steering_torque=1800 > Change the torque to 2000 and then 1500 and try those as well. Let me know > how it works on your wheel.

Richard G Clegg <[email hidden]> wrote in message news:89gc1i$e2f$[email hidden]... > As I understand it, the "shimmy" you describe is a result of having > the latency set too low. As I understand latency it is a measure > of "how far ahead you want to predict the wheel setting" - you can't > get the "force" to the wheel instantly, there is a certain delay. > Therefore, GPL predicts what the force WILL be in a certain amount of > time and sends that to the wheel in the hope it will get there. I'm not sure that I agree. While I still don't understand how the numbers change what, setting a much higher damping value cured this. Looking at the effect, I would guess that this value is properly named: under-damped systems oscillate; higher damping fixed the problem. I agree with you about latency, though, since my reaction to any delay is probably typical of everyone else. :-) It seems just as driveable at 0.01 as 0.05. I haven't tried values outside that range. Michael.

Michael Young <[email hidden]> wrote: : Anyway, I've since tried changing the FF numbers to see if I could get a : lighter feel without losing too much feedback. I never did find a good value : to use for max torque. It would either shimmy strongly, or was too light to : be of much use. By shimmy, I mean the wheel would slap back and forth if I : let go, even at slow speed. At one point, I could distinctly smell the ozone : of an overstressed electric motor. I realize that the shimmy is an artifact; : not finding resistance, the aligning forces would quickly turn the wheel : until it reversed, and then it would feedback the other way. However, I : never did find a balance that I liked. As I understand it, the "shimmy" you describe is a result of having the latency set too low. As I understand latency it is a measure of "how far ahead you want to predict the wheel setting" - you can't get the "force" to the wheel instantly, there is a certain delay. Therefore, GPL predicts what the force WILL be in a certain amount of time and sends that to the wheel in the hope it will get there. A lot of GPLers simply assume "low latency is good" (right) and therefore set the latency as low as possible (wrong). As I understand it, if you set the latency too low, once the force gets to the wheels, it "lags" what the wheels are actually doing. Imagine you are travelling down the straight, your wheel is slightly to the right - there should be a restoring force to move your wheel to the centre. Therefore, GPL sends a "move wheel left a bit" message to your wheel. However, if the latency is set too low in GPL then by the time the message gets to your wheel and moves it left your on-screen wheel is straight.... as I understand it, this rocking from side to side is a problem of low latency setting up a non-linear dynamic system (to borrow a phrase from my work). Try popping the latency up a bit higher and see if it helps. Richard. Disclaimer: other people on this group probably understand the FF settings much better than me and might find flaws in my reasoning. -- Richard G. Clegg Only the mind is waving Networks and Non-Linear Dynamics Group Dept. of Mathematics, Uni. of York www: http://manor.york.ac.uk/top.html

Richard G Clegg <[email hidden]> writes:

Quoted text (33 lines)

> Michael Young <[email hidden]> wrote: > > : Anyway, I've since tried changing the FF numbers to see if I could get a > : lighter feel without losing too much feedback. I never did find a good value > : to use for max torque. It would either shimmy strongly, or was too light to > : be of much use. By shimmy, I mean the wheel would slap back and forth if I > : let go, even at slow speed. At one point, I could distinctly smell the ozone > : of an overstressed electric motor. I realize that the shimmy is an artifact; > : not finding resistance, the aligning forces would quickly turn the wheel > : until it reversed, and then it would feedback the other way. However, I > : never did find a balance that I liked. > > As I understand it, the "shimmy" you describe is a result of having > the latency set too low. As I understand latency it is a measure > of "how far ahead you want to predict the wheel setting" - you can't > get the "force" to the wheel instantly, there is a certain delay. > Therefore, GPL predicts what the force WILL be in a certain amount of > time and sends that to the wheel in the hope it will get there. > > A lot of GPLers simply assume "low latency is good" (right) and > therefore set the latency as low as possible (wrong). > > As I understand it, if you set the latency too low, once the force > gets to the wheels, it "lags" what the wheels are actually doing. > Imagine you are travelling down the straight, your wheel is slightly to > the right - there should be a restoring force to move your wheel to the > centre. Therefore, GPL sends a "move wheel left a bit" message to your > wheel. However, if the latency is set too low in GPL then by the time > the message gets to your wheel and moves it left your on-screen wheel is > straight.... as I understand it, this rocking from side to side is a > problem of low latency setting up a non-linear dynamic system (to borrow > a phrase from my work). Try popping the latency up a bit higher and see > if it helps.

Thanks for the tip. I've always known why the wheel go wild, but never how to fix it. I'll try it when I get home from work. -- Olav K. Malmin remove spam when replying

Oddly enough, my best F1 time at Monza was while I was feeling quite relaxed (and a bit dizzy) ;) Try these settings, what have you got to lose? Damping=4000 Steering_torque=1800 Change the torque to 2000 and then 1500 and try those as well. Let me know how it works on your wheel. Slot "Michael Young" <[email hidden]> wrote in message news:[email hidden]...

Quoted text (9 lines)

> GTX_SlotCar <[email hidden]> wrote in message > news:Ctvu4.292$[email hidden]... > > What settings are you using now and what did you use before? > > I'm using a latency of 0.010, and a damping force of 35.0. Max torque is > commented out. I'm looking for best feel for road conditions, without > inviting repetitive stress injury. > > Before, the forces were high enough that my hands and wrist were still

sore > the next morning. Two weekends ago, I ran a full GP at Spa plus the second2 > hr practice before that, about 3 hours of actual track time in onesitting. > (I finished 5th, and was pleasantly pleased, if tired and drained. I had > only recently made some strong breakthroughs in lap times.) I thinkdamping > force was 85, and possibly the initial default 0.035 latency; max torquewas > still commented out at the time, I believe. Part of the problem is Ididn't > make note of how it was set before I started making changes (who wouldthink > you would so easily forget 3 little numbers?). The wheel was heavy, and > kicked strongly from time to time; road feel was very good, and the shaker > let me know when I was running right out to the pavement edge. > > Anyway, I've since tried changing the FF numbers to see if I could get a > lighter feel without losing too much feedback. I never did find a goodvalue > to use for max torque. It would either shimmy strongly, or was too lightto > be of much use. By shimmy, I mean the wheel would slap back and forth if I > let go, even at slow speed. At one point, I could distinctly smell theozone > of an overstressed electric motor. I realize that the shimmy is anartifact; > not finding resistance, the aligning forces would quickly turn the wheel > until it reversed, and then it would feedback the other way. However, I > never did find a balance that I liked. > > So, that's why I'm asking. Last weekend, I thought I would amuse myself by > taking a break from struggling around Rouen, and take a few turns aroundSpa > while loaded up on a half bottle of Merlot (Lynfred Winery 1999 Private > Reserve; absolute nectar if you like this kind of thing!). My second lap > out, I turned a PB of 3:21, 0.8 seconds faster than ever before. It was at > this point that I started to monkey with the FF numbers, and thingshaven't > been right since. Even if I can't improve on it, I'd like to at least get > things back to where they were. Hence my appeal for your help. My laptimes,

Quoted text (7 lines)

> at both Rouen and Spa, are now almost 2 full seconds slower than at that > peak of alcoholic bliss. The only changes were to the FF settings. > > Michael. > > >

"GTX_SlotCar" <[email hidden]> wrote in message news:b%Zu4.5$%[email hidden]... > Oddly enough, my best F1 time at Monza was while I was feeling quiterelaxed

Quoted text (10 lines)

> (and a bit dizzy) ;) > > Try these settings, what have you got to lose? > Damping=4000 > Steering_torque=1800 > Change the torque to 2000 and then 1500 and try those as well. Let me know > how it works on your wheel. > > Slot >

What settings for MSFF do you use in control panel?

On Mon, 28 Feb 2000 20:52:43 -0600, "Michael Young" <[email hidden]> wrote:

Quoted text (7 lines)

>GTX_SlotCar <[email hidden]> wrote in message >news:Ctvu4.292$[email hidden]... >> What settings are you using now and what did you use before? > >I'm using a latency of 0.010, and a damping force of 35.0. Max torque is >commented out. I'm looking for best feel for road conditions, without >inviting repetitive stress injury.

My original Papyrus settings were: [ Joy ] allow_force_feedback = 1 ; Use FF if device has it force_feedback_damping = 25.000000 ; force feedback damping coefficient (org) force_feedback_latency = 0.085000 ; force feedback latency (secs) org Sierra max_steering_torque = 225.000000 ; steering torque in N*in giving max device force Never played around much with them though. A bit more damping, a bit more torque, but didn't really test much. Ruud van Gaal MarketGraph / MachTech: http://www.marketgraph.nl Art: http://www.marketgraph.nl/gallery

I would set the overall device gain to max, and all others to 0. Slot "Bart Westra" <[email hidden]> wrote in message news:89icc3$s9a$[email hidden]... > What settings for MSFF do you use in control panel? > >