Separate names with a comma.
Discussion in 'Assetto Corsa' started by Thom Lee, Apr 28, 2014.
This video helped me with driving Nordschleife
Thanks so much for your time and help.
About the 2nd, I have zero feedback on straights, even on very high speeds. I use G29. Should I change my feedback settings?
@ArchKyuubeey great info and videos. Pedal box videos are especially helpful.
I'm viewing my replays to evaluate myself. I see some small lockups but I can't feel them during driving. I may have problem with feedback settings.
Edit: FB settings are currently set as below:
You don't have feedback dead center because the G25, G27, G29 and such use opposing gear setups to transmit forces. If you would get forces in center, the gears would fight each other and oscillate. Go for quite low FFB in these G-series wheels, like 50% gain in options. A true direct drive wheel will transmit forces basically dead center, SUPPOSEDLY without a real deadzone but I don't know personally.
Locking up with steering straight won't be felt via the wheel in a G-series setup, unless perhaps if you enable vibration from locking up. I don't think I have it enabled, nor do I know what it really does. I suspect it vibrates the vibration motor when you lock up. Would be cool if it was in the brake pedal.
portakal you need to tick the enhanced understeer effect, you really get a feel through the wheel when front grip begins to go away.
Thanks. Do you use the LUT generator on RD? I just saw a recommendation on Reddit for it.
Sorry for basic questions everyone.
Edit: I'm researching and trying FFB setups. Thanks for your time everyone. This community is awesome.
Turn Gain to 50 - 60% or so and check. It should be stiff with feedback but should preferably not clip at all, or a little. I run quite low, near the midrange.
G-series wheels will not last 8 years if you run 100% I suspect: but mine has lasted 8 years to now running 50% and still strong .
Filter OFF completely. Never ever on this kind of wheel IMO.
The additional settings for kurbs etc. are user specific but I use 0% on all. Perhaps I will experiment.
This got me thinking (I forgot all about the FFB settings in the driver), so which gain do you use, the driver %FFB gain or CM's %FFB gains?
Or does CM's %gains over ride the driver settings?
Or any of the settings for that matter. I havent looked at this page in eons.
You set the driver specific FFB to 100%. In some wheels it will affect the gain curve, so having it under or over 100% will produce unwanted results. It's independent of the sim's gain, as the sim's gain is the output coefficient.
Turn Spring and Damper to 0% if you're on a G-series. If you have a more sophisticated wheel, perhaps some damping like 10% might emulate a more realistic feeling, but I didn't feel I gained much from it on my G-series. I suspect right now you have a constant spring centering force BUT there's also a chance sims just override it, hopefully. Unless "Spring" doesn't mean what I think. In my wheel, it's a constant self centering force that's not dependent on anything.
OK cool TY. That makes sense, if its filtered by %25 before it gets to CM, its not getting %100 of the signal on the output....never thought about it.
Yep. That's essentially what the global gain will do. I don't know for a G series if it's say, 50% gain ingame and 50% gain global = 25% of input, or if it actually touches the curve. I remember seeing some curve data and even LOGGING MY OWN curves, but I can't tell you for sure. Was so long ago.
I have a hunch Spring does nothing in AC except perhaps pull a little in collisions, actually. Especially if self center is at 0. Damper should basically just be, well, a damper. Mainly you'll feel it when countersteering, it'll push back a little if the sim supports it.
100% gain, 50% ingame, 0% for canned effects ingame (They're already present to a degree and I suspect this is just amplification) and 0% for Spring and Damper works fine for me. A lot of real PS cars are lighter than most Kunos cars at 50% ingame gain; I try to account for that myself via the car's FFB strength in the data.
EDIT: Oh and the G27 at least is very dependent on temperature. It'll not be noticeably as strong when cold as when warmed up.
Maybe if you look at the tyres file, some front Radius is wrong...
I did experiment a bit based on your suggestions. I've upped the gain to %70. Road to %2, kerbs to %5 (trying to replicate F1 2019 feel with kerbs). Game feels great now. I should have done this before lol.
Also I was still getting no FFB on center with these settings. Creating a LUT file with WheelCheck and LUT generator fixed it. Now I get road vibrations as well.
Don't try to "fix" the FFB dead center. As I said it will result in the motors oscillating and often clipping.
I downloaded one from the link accumulator that is AssettoLand, perhaps you could attach the cars 'data' folder for me to look at?
Fredrik noticed what a lot of real-world drivers have noticed about AC. There's too little front grip as you start to brake (and weight shifts to the front), then the car shifts to too much oversteer when you get back on the throttle. It's undoubtedly why drifters love AC. The issue varies from car to car, and a skilled modder can (apparently) tune it out to a significant degree.
It's also probably why people who only play AC think that "lift understeer" exists. In the real world (and in every other sim, even Forza 7), no such effect exists. In over 30 years of driving cars at the track, I've never once lifted mid-corner and had the car lose grip at the front. Do that in a real car (or most any other sim), and the car will rotate (and maybe spin). I feel stupid even having to say this. It's like arguing whether grass is green. Lifting to tighten your line through a curve is such a basic, fundamental aspect of performance driving that, for anyone who's driven at the limit, it's second nature.
AC's handling of curbs is also quite messed up. IRL, curbs are your friend (with some notable exceptions). That's why we call it "curb-surfing". But try that in AC and you'll be constantly spinning out. This time, it's a known, acknowledged bug and is mentioned by IER in their excellent Porsche GTA mod:
"- Car is very bad over uneven apex surfaces like curbs (tire model issue with AC - worse on some cars than others). Exit curb behavior is okay."
AC is a fun game with so much variety (why we're all here, right?), and it has great FFB. I wouldn't use it for driver training. Oddly enough, I really like Raceroom (or any other gmotor-based game) to give me that "almost like I'm there" feel. But if you get the right mod, AC is very close and the FFB is second only to rF2 (imho).
There is no one perfect sim out there. Each has its own issues vis-à-vis Real Life.
With good load curves, you can basically completely tune out that behavior. In addition, high-aero cars with vanilla tires have wayyyyyy too much grip to a very high degree.
Have you really never been at apex, lifted off to get more turn in and pushed because you scrubbed the fronts, then had it whip back on the exit when the fronts grip up? It's the most common driver mistake in all driving apart from maybe turning in too early. If you have some footage online, I'd like to see it, because I'm not sure I understand the context of your driving.
EDIT: Oh, and curbs are messed up. Even KS admits it in ACC.
Any suggestions on what tyres I should use for Group C cars with over two tonnes of downforce?
You need to make your own load curves for them. Nothing with KS equations will be even remotely close. Perhaps take a look at the NP35.
No, no, no. The only time I've seen what you described is when someone panics and is simultaneously trying to brake deep into the corner at the same time they've got the wheel cranked. Eventually, they scrub off enough speed and then suddenly the fronts grip up and you get a tank-slapper.
This is what you're trying to describe:
Driver enters a turn too fast, on the wrong line, or some combination of both;
Driver *slams on the brakes* and turns at the same time;
This completely overwhelms the front tires, which lock up and skid along uselessly for a while, producing instant and total understeer;
Until the car slows down to the point that it's possible for those front tires to grip;
Which they promptly do;
And since the driver still has the steering wheel cranked, the car steers sharply without warning;
Which slows the front end of the car further, because you can't get a lateral motion of the car without sacrificing forward motion, because there's no free lunch in physics;
Thus producing all the conditions you need for a rapid spin.
The key thing you're missing is the brakes are the real culprit here.
That's not at all what I'm talking about. I am not on the brakes. All I've done is lift off. The rear gets light and the car rotates (a lot or a little, depending on a million factors but mainly where the engine is and how tight the diff is). I have never once had a car "scrub the fronts" simply by lifting off. For that to happen, weight transfer alone would have to be enough to overwhelm available front grip.
Nowhere out there will you find the term "lift understeer" or "power off understeer". Why? Because simply lifting (and thereby causing weight transfer) isn't enough to overwhelm the traction on any reasonably-sized front tire. You need to brake simultaneously to do that. Sure, if you put a motorcycle tire on a Mustang, this idea is theoretically possible. But no car (outside a drag car) is designed like that.
And a million other places. Nowhere are you going to find a driving instructor talking about inducing understeer by shifting weight forward alone. Again, you have to exceed the front tire limits by braking and turning at the same time to do that (on any reasonably sized tire setup).
When you're talking about oversteer, how much yaw are we talking about here? It's not impossible for delta yaw to become lower by lifting off in the right conditions. Although you are right in the general sense that if you *kept* off the throttle with the same steering angle; eventually you'd probably get some oversteer, or more neutral steering to be precise.
Almost no production-based car is actually gonna oversteer only from lifting off on say, a dry skidpad, just go away from the extreme understeer bias present in production cars and most modern racecars towards a more neutral handling: but it's still understeer. Not every car is gonna whip around and do a 90degree turn suddenly, unless perhaps you're not yet set and then you just jump off the gas. You're probably thinking about not yet being set or being in transient and then just jumping off the gas.
Actually it'd be pretty interesting to just empirically measure this. You'd be surprised what actually happens to the yaw moment and especially yaw vs slip direction. I'm thinking it in the sense that a car which is drifting is *heavily understeering*; you would probably call it oversteer. Makes more sense?
I have a 2017 MW M6 currently, and have previously owned a 2006 M6 and a 2003 Honda S2000 (plus a lot of older cars). All of those exhibited snap-oversteer at the limit if you suddenly lifted - especially if you're in a lower gear (sudden decel/weight transfer). All cars exhibit this to some degree.
OTOH, my 2013 Audi TTRS was much less, shall we say, adjustable. Yes, it would tighten its line if you lifted, but it was way more subdued. And even with all driver aids disabled, the car was simply un-spinnable (using throttle alone - you can spin any car by being a bonehead). I drove a Miata once and it was pretty gentle as well.
Look, I'm not talking about someone's Camry or base-model Golf. I'm talking about sports cars. Those are not set up to understeer no-matter-what. If they were, they'd drive like crap and get bad magazine reviews. I have this feeling maybe you're extrapolating from your own experience in low-powered FWD sedans or something. I can assure you that something like an M6, S2000, 911 etc is not set up to drive that way. Those cars really need stability control for normal people to handle the car on a track. How do I know this? Because I've seen plenty of rich dentists and plastic surgeons spinning their 911's, McLarens etc.
Usually they do it by suddenly lifting off when they enter a corner too hot (plus often smashing the brakes at the same time - see my post). Modern cars are way better about this, of course. But that stability control is there for a reason. "Never lift" is something you will hear a lot if you talk to drivers of older sports cars in particular. The S2000 (which had no stability control) was absolutely infamous for this. Just google "S2000 snap oversteer" and get ready for about a million hits.
Just so you know I'm not BSing you, here are some pics of the cars:
I think this is more an issue of my view which is based on what is *really* happening physics wise, and your view which is the driver's perspective. I'm not sure we're *actually* disagreeing.
Let's say it like this:
You're coming in hot for a corner in an old 911, you trailbrake without locking up at all, aannd now the car's going sideways. The car will do a 180 if you don't do anything: you're oversteering. You countersteer a little and start getting on the gas lightly.
You're not yet countersteering but steering is close to center, and the car's at, say, 10 degrees angle of attack sideways and you're modulating throttle to keep it steady. You're probably slightly understeering, close to neutral steer at this point.
You countersteer, get on the gas and start to drift. Your lateral angle of attack is high, as high or higher than when the car started to let go, but your yaw amount has decreased. You're now tracking a wider line than at apex, unwinding out of the corner, accelerating almost completely straight out of the corner despite the car still being angled. You're still understeering.
See what I mean? Even an AP1 S2000 will inherently understeer or at the very least get close to neutral steer; unless perhaps you steer in quickly and jump off the gas at high speed. The basically nonexistant coast lock + very high rear lift bias + very neutral suspension means it'll probably start to oversteer very soon after some initial push and neutral steering.
There's a chance you won't even countersteer in time and you will indeed oversteer so much that all four start sliding excessively and you start going right off the track if you screw it up bad enough.
You can test it pretty easily, just go to anything resembling a skidpad and see if you can drive at the tires' limit and track a radius. If you can, congrats, your car understeers. If it keeps wanting to rotate and you keep decreasing steering or throttle just to stay on the line: your car's very neutral. If it oversteers, you won't track the line for longer than what it takes for the car to decrease radius enough to steer itself off the line.
The amount of "instablity" or "neutral handling" or "oversteer" a car exhibits is abstracted into the dynamic index, and when your driver says the car "oversteers", he probably means "it's not understeering enough".
PS: I made MY00 AP1 physics which was very well received by owners, and it did match up to footage very well, so I have a bit of an idea how AP1's drive. For a motor vehicle company it's quite daring to release something like that to the public.
Damper must be set to 100.
AC uses damping in FFB. If you clean it up, you won 't get all the effects in the game.
This is an important effect for simulation.
Unfortunately, this is a feature of Logitec G-series rudders.
In the middle, the steering wheel is always empty. To fix this, it is necessary for the developers of the simulator to write a separate FFB specifically for these rudders. But that 's not what anyone 's gonna do...
you need the updated config for it from github, included one is broken:
(right click "raw", save as...
copy to "assettocorsa\extension\config\tracks\bugatti.ini"
delete any "assettocorsa\extension\config\tracks\loaded\bugatti.ini")
I've seen the devs for both AC and rFactor say it's best to keep the profiler gain at 100% and then adjust the in-game gain to taste. They both said the theory that you need to keep it below 75% or whatever isn't really accurate or necessary. Since reading that I always keep my profiler at 100% and adjust in-game as needed and I have no complaints.
For those wanting to reduce center deadzone in Logitech wheels you might try the LUT generator.
What is Damper used for in AC specifically? Perhaps I'll turn it on and see. To my understanding it's just essentially some "stickiness" when countersteering: do they use it for anything else?
For sure, a real wheel setup compared to say, a "clean" DD sim wheel is A: Lighter B: More damped than the usual setup.
I believe Stefano (AC dev) also said that damper doesn't do much in AC, adds a little weight at a standstill or very low speeds but once you get over about 5mph it doesn't do anything. And of course keep in mind that your in-game damper settings can only go as high as your profiler settings, i.e. if you have your profiler at 50% and the game at 100% you're only getting 50% damping because that's all the profiler is allowing. Profiler at 50% and game at 50% would be 25% damping (50% of 50% = 25%), etc.
It gives a viscosity.
With this effect, the weight of the car is modeled. The most notable application is when the car is standing, the steering wheel is hard to spin.
Turn off the damping in the profile, and the steering wheel on the standing car will be empty. But not only does this give a damper, just another influence is harder to notice in motion.
When AC first appeared, there was damping adjustment in car setup. It was very convenient when the steering wheel seemed too light, empty. You could add a little bit to increase the feeling of the weight of the car. But in the updates, it was removed from the settings.
I think this parameter can be set in the configuration files of the car, but I don 't know where. Otherwise I would add a bit more to some cars - FFB becomes more realistic, and more pleasant to drive.
Hello all , I have a question about tyres :
245/645-18 = 245/40-18 ?
305/645-18 = 305/40-18 ?
In assetto corsa tyres.ini :