Hey F3 experts, what's the best way to get some low-speed rear traction improvement on this thing? I have been playing with diff settings with some useful result, but what do the gurus suggest?
No expert but... Lower rear anti roll bar does wonders for rear end stability O and I very rarely use 1st gear
Thanks. As usual, it's 95% skill and 5% set-up. But the rear ARB does work nicely to affect this. The joke is I set it as high it would go and ran a bunch of laps and then set it as low as it would go and did the same. I found I could go fast with either extreme if I just paid attention and adapted properly to the car. But in general, lowering it is a successful way to settle the rear on sharp, low speed corners. You are more expert than you think! And yes, I don't think I have used 1st gear on the track ever in the F3. True for almost all race cars and tracks. 1st is just to get the car moving and then you never use it again until the next time you need to launch. With most sequential transmissions, even paddle shifted ones, you still need to use the clutch to get into 1st gear (either at launch or on your way down from second). That's reason enough to avoid it
Also, try and get familiar and as comfortable as possible with the physics engine and how oversteer acts in this (or any sims') physics engine since they're all different. Unfortunately, you must try to stay under the limit more or drive with constant lazer-like accuracy with the oversteer in the Reiza/ISI/rFactor physics engine rather than having the "playfulness" and adjusting of wheelspin, opposite lock, and oversteer angles like possible in real life (and some other sims especially Live for Speed). It's much more on/off - and with very quick, short-lasting saves, slides, and angles in this physics engine. There isn't that "middle area" during the oversteer with the control and manipulation of the slide allowed like in real life (or, again, Live For Speed as the prime example). Basically, don't only adjust setup because even with a softer, grippier rear-end, you'll still eventually hit the limit of rear grip (unless at every corner you're always understeering way before oversteering which is difficult to do unless you setup your car to drive like a truck). You just have to try to get an idea and feel for what each games' physics engine likes and then try to stay within those areas. Think of what the games' physics engine likes, not real life.
Don't agree. The AMS cars do not in general suffer this failing that many of the rFactor and rFactor2 cars do. The F3 is controllable, even after you start to spin if you moderate the throttle properly and help out a bit with a few set-up tweaks.
It is controllable, I never said it isn't . Still suffers from the behavior though. It's inherent to the physics engine. It's easily visible from onboard views, from TV/trackside views, from driving with FFB, without FFB, high-end wheel (eg. OSW), low-end wheel. Of course it can be controllable if you stay within the sweet-spot or drive accordingly. Having a car that looses 500% grip at anything under 100 Km/hr (completely made up example, nothing to do with AMS or any sim) can still be controllable if you understand that behavior, avoid it, or get used to it and learn to drive within those limits. The controlability that can be had over a particular driving characteristic doesn't mean there isn't a problem/flaw with that particular driving characteristic. Not to mention, it's clear from any real life rear-slide in any racing series (from F1600s to F1s) that the behavior is quite different.
Marc, do you know these cheat sheets? They helped me a lot, especially the turn in/out/apex parts as well as the diff parts in the 21 steps guide. I loaded them up for you: KillKenny - a safe home for all your data Maybe you cloud have a look at my Adelaide '16 F301 TT setup, however I tend to be slower on someone else's setup so I don't know if that's helpful.
Thanks! I have a few set-up guides and books, of course, but this one is very nicely organized and easier to comprehend at a glance than most. I'll check-out your set-up, too, but as mentioned above, playing a bit with the ARBs solved my issues. You can dial the car pretty much from those and maybe a diff tweak.
I have spun a Skippy and a Radical SR3 at comparably low speeds. In fact I only spun those cars at low speeds, not at higher speeds. It happens when there is too sudden a shift of weight combined with either a slippery patch on the track, too much throttle application, too sudden throttle application, too soon throttle application, see-saw brake application while turning, etc. You know I agree with your premise that many ISI-based sim cars suffer from too light a rear end. I can't say that I know exactly why, but it is true. What I can also say is that some cars do not have this problem. The rF2 USF2000 is one. Almost everything in AMS are others. AMS has configured their vehicles differently from the old rFactor ISI cars and many in rF2 (that were probably modelled after the old rF1 cars). They are more consistent to each other and overall feel like they all belong in the same physics universe and that universe is the same one I live in and have driven in for 30+ years. No other sim gives me that experience, including iRacing and rF2 that, while they are great in many ways and superior to AMS in some ways, both suffer from inconsistency from one car to the next in terms of physics and FFB design (my guess is due to people relying on engineering data to the exclusion of seat of the pants reality checks). So, sorry, but I am still respectfully disagreeing with your blanket conclusion.
Btw, one thing that I have overlooked for quite some time are the packers, because I never thought they would have such a high impact on the car. Try to reduce them a little bit
Any time you get snap effects when turning or braking, it could be the car riding on the packers. Can't remember which car it was (not F3), but one was acting almost like it had no suspension. Raise ride height a bit and lower packers and voila, a normal car all of a sudden. I think it has been fixed in the meantime.
I agree that AMS cars are much more consistent across the field and I generally agree with everything or most of what you said but - no disrespect intended - what you wrote in your post shows you aren't comprehending what I said in my previous posts (and about the ISI engine in general). You can drive a car with a super-light rear-end in any other sim or even real-life, it's not so much about that but about the vehicle dynamics/kinematics on how the overall physics are behaving during those situations which is still quite in need of repair on the ISI engine. It's the overall behaviour of what happens during the oversteer which is quite different to real-life regardless if the car's rear is difficult or easy to get loose. Reiza did make two updates that slightly improved the situation: 1. Some sort of inertia update a year or so back - I have been talking about this issue for years 2. Some engine (as in "motor" engine) updates rergarding how torque loads work relative to throttle - I have been also talking about that issue for years I have known about those 2 abpve issues (and more) for around 10 years now. I've been quite vocal about both those issues (and more) for at least 5 years now. I'm often flamed or argued against in forums...Then, in the last year or so, Reiza of all people (not ISI) applied some improvements to those 2 issues I always tried to tell people about for 5-10 years. Maybe people should start listening to me more
I comprehended it, I just don't agree with your blanket statements about AMS. I do agree that the issue you refer to is real and legitimate (and been around for years). The changes Reiza have made solve the problem for me. ISI/S397 have released at least one car that doesn't have the problem. So my conclusion is that it is not the engine itself that has the problem, but the implementation of it (car modelling). In other words, don't blame the tool when it is the user who doesn't know how to use (program) the tool properly.
I've never driven a car or even seen a video of a car running on an ISI engine not prone to this since I started simracing in the year 2002 (with F1 2002). The way the rotation of the rear end changes and slows down or stops as you apply more opposite lock is quite different to real life. In the engine, cars often, out of nowhere, stop sliding as if inertia (or weight or momentum or whatever) suddenly seizes to exist. Also, you can often see that the slide ends from the car ending it's previous trajectory of travel while the car only partially rotates/corrects back to centre; in other words. the vehicle only partially rotates back in line in conjunction with the slide just sort of vanishing out of nowhere. This also makes big angles relative to the vehicle's trajectory very unrealistic as, upon correction, you often have to return the steering back to centre in a somewhat premature way and almost just hold the steering near centre (maybe just a few degrees off) and sort of wait for the slide to end or else the car will suddenly snap over-correct due to my previous point about slides suddenly ending unnaturally. This is in stark contrast to real life where the car almost always keeps heading towards it's continued path before the slide occurred in combination with the driver giving the steering a nice "boat-load" of opposite lock all while the car's rotation slows down and settles accordingly with the opposite lock applied - it all flows completely naturally with each other. Connected to all this in-game behavior, also has an effect on turn-in or mid-corner. No other game-engine, or in real-life, allows you to make the car arc further into a sharper turn by simply dragging the brake or getting the rear-end loose like in the ISI engine (sort of like a trail-braking "hack"). Often, you don't even need to add more steering lock or adjust throttle, instead, you just keep your steering and throttle static and kind of give light pumps to the brake or keep dragging it in order to make the car almost "magically" arc in further and therefore turn in more. In real life, instead, the rear will just get loose and want to rotate as the car holds it's forwards trajectory. An obvious (but certainly not limited to) example is on oval racing: in real life, when the rear rotates, the rear wants to simply rotate around while the car heads outside towards the wall for that moment, but in-game, you can use that rear-spin to simply make your car turn-in sharper and sharper since the car's trajectory wants to follow where the car rotates to rather than traveling on path much more independent of the rear-rotation like in real life. So instead of having the rear-end come around relative to the direction of travel, you just get this ultra turn-in arc effect. Real fast players really know how to exploit this oversteer behaviour and I've been seeing them do it for 15 years in the ISI engine. This also often leads to hardly any lock being needed to get cars around corners when exploiting this behavior; I could literally get Nascar cars around Indianapolis in RF2 using like 20% of the steering lock real drivers use and I was using an even slower-than-default steering rack. I would barely turn the wheel or play with the throttle but instead just fly into the corner then play with the brake as needed to make the car turn-in (not oversteer relative to the vehicle's trajectory.but simply turn-in) extra amounts. I firmly believe this super turn-in arcing phenomenon is all related to the oversteer correction stuff I first talked about. They're all by-products stemming from the same core-issues as they all have very similar behavioral connections. Moving away from turn-in and mid-corner and back to corner-exit, this is also why so many more spins in the ISI engine end with the car doing either a sharp turn and simply driving into a wall (as if the car suddenly turned 90 degrees with super grip) or with the car snap corrected in what looked like a totally under-control slide. It's to do with the slide suddenly, instantly ending (as if inertia suddenly, momentarily seizes to exist) and with the vehicle's trajectory (direction of travel) "fusing together" with the slide rather than the rear-end sliding-out and correcting back independently of the vehicle's trajectory path. I want to stress that Reiza have made improvements to this in 2 areas which I've been calling on improvements for 5+ years in the form of their inertia and motor-torque updates.
