Automobilista 2 V1.2.0.0 & Racin´ USA Pt1 RELEASED - Now Updated to V1.2.1.4

Just checked out how the Mercedes AMG GT3 works with real world values.
front axle 0.2°, rear axle 0.4°, Brake 60/40, LSD 60/40, 6 clutches, 80Nm preload, TC3, ABS6... works pretty good and stable.

I don't undertand the use of the current default values

 

Please consider this post my attempt to hit the agree button 100 times. I can learn how to drive different cars. I want the racing sim aspects completed.

 

Ok for starters, you are not totally inducing pure power oversteer. You can see in the video you are throwing the balance of the car off by moving the wheel back and forth. For example at turn 2.

I did your same settings (as best I could, how do you still have actual values in your setup whereas I just have generic numbers?). With that said, I did get a slight bit loose with power oversteer on hairpin T1. After that, nothing. But just to get to that, you had to get to your setting which are a bit extreme, including lowering wing, and then set TC to 0. So you have to extreme settings, TC off and then you can barely create a bit of power oversteer. That is a problem.

 

All of these words, but you completely ignored I posted the Merc GT3 video with the exact same problem. And that's on a car you say has a "power oversteer character"

As for the Porsche, here is the Boxer Cup from AMS1. Default Setting and TC Off. Hockenheim T1 out of the pits and I immediately power oversteer, very similar to ACC.

So is AMS1 wrong too?

 

I have my Time Trial at the default settings, which means Track Setting is "Default Progressing". So unless there is a bug, it should not be at Optimal setting to start.

In either case, even at optimum setting, all sims share this power oversteer characteristic as ACC. What about Assetto Corsa, Project Cars 2, Iracing, AMS1? They all have the same characteristics as ACC.

I just posted the AMS1 Boxer vid above. Did Reiza get it wrong before?

 

Haha, I feel the same. I've posted like 10 vids showing the problem and people still keep arguing that's not a problem or that's how it is supposed to be when no other game maintains that behavior. I have to be honest, I'm not the one being stubborn here. I love playing this game, support it on other forums and with my wallet and want it to be the best sim out there. But if there are improvements needed, we should recognize them rather than pretend they aren't issues.

As for longitudinal grip as a specific work in progress, where did you get that information? I have not noticed that mentioned. The closest I've seen is that TC was still being worked on but I think those updates were all just released and it has not solved the problem.

 

If you like to spin and "donouting", pick sprintrace and have fun.

 

best for that sole purpose is the McLaren P1LM Street car.

 

whatever McLaren you can find under "Street Cars"...last time I checked there were only 3, and onyl one was a McLaren. Hard to miss it

 

You guys are all arguing over an unfinished product. Reiza are working on it. They've never claimed this is how the cars should drive. They have only ever stated that they are working on it and fundamentally things are gradually getting more and more realistic. I jumped into AMS 2 yesterday to see just how bad the physics are ended up having hours of fun.

My only gripe at the devs is that they haven't updated the forum's FFB setup guide. FFB setup is so critical in AMS2. Take Racerooms in game FFB setup for example.

 

I know this discussion already ended, but I wanted to address the notion of extreme setups further. It’s actually defaults that are extreme understeer settings in many cars. Way softer rear ARB than front ARB already makes it so that when weight shifts to either rear corner on throttle, front does not grip well. I’m basically parroting @steelreserv, but it’s worth repeating.

Power ramp of 78 is on the loose side in the AMG. 2-4 clutches lacks a lot of overall diff friction. Overall I guesstimate that this is like having 5-10% power lock in other sims.

Also if the setup can induce power oversteer, you will notice the start of it TC or no TC. (Unless you’re at 10)

The fact that you have to make extreme changes relative to the extreme default does feel wrong, but it actually tells us that setups can make a big difference through physics. The stubborn power understeer phenomenon discussed here is mirrored in BeamNG with similar suspension and even more generous diff settings in an RWD supercar (Bolide) on slicks.

But at the end of the day I agree with your original assessment that out of the box there is no power oversteer with the cars you mentioned. It’s just that the nature of the problem is not damning compared to some deeper root cause in the back of one’s mind (Madness bad etc). And so it has less to do with physics and more with setup changes that coincidentally prove that the physics work. Although the slight lack of longi slip was also mentioned. Even so the default setups deserve reviewing, because that’s where most of the issue comes from. Beating a dead horse .

There are also a few cars we can’t setup to our liking that practically don’t have power oversteer. These are frustrating to me.

 

Several things can be true at once:

1. The Porsche 911 is designed for maximum rear grip out of corners and exhibits on-power understeer as a rule.
2. The Porsche 911 GT3 R/RS still have enough hp that if you turn TC off and really stamp on the throttle with full steering lock applied, it should be possible to break traction.
3. Diff settings of several cars in AMS2 are still a bit too open to ameliorate some earlier drivetrain issues.
4. GT tyres in AMS2 might be a bit too gradual in the way slip and grip occurs.
5. Other sims using variants of the empirical Pacejka- tyre model exaggerate the ease at which grip is lost at low speeds.
6. The Seta tyre model can be used to make tyres as grippy or as slippy as you'd like and there are no inherent limitations native to the engine.

 

Evolution not revolution has to be the way forward to address issues like these. In that context an acknowledgement that a problem exists is a necessary precondition to resolving them.
That has happened.
Time to let the evolution bit happen..

 

4. Sounds like a good summary to me.
5. ACC used to be terrible with this but they have really improved this over time such that I don't notice as much any more. Same goes with rf2 I guess.
6. How do you know this exactly? How can you be sure there are no inherent limitations just like there are with the old Pacejka model? Not being smart, just curious.