Physics discussion thread

Is your My Docs folder synced to Onedrive Cloud? Maybe you have a controllersettings file corruption happening that's messing up your controls for AMS2? Maybe try disabling Onedrive Sync for your My Docs folder?

I know I stopped syncing my My Docs folder to Onedrive and stopped having weird controller problems that were plaguing me.

 

From reading I am not even sure where the issue is to be honest, because the info provided is not really accurate and there is no video of what the actual inputs show. So might be a "physics issue" because of how the throttle response is in AMS2 or just hardware/settings related.

I do agree however there are some cars/classes that lock up pretty easily compared to other sims, I often have to run with heavily reduced brake pressure as a consequence but I do not whether this is realistic or not.

 

The tyre wear is set to "authentic"

Should it be something else?

 

No, and I won't.
The most recent reason appeared after the patchnote, I had 2 of them.

 

Hello guys.
Love Ng time ago I discovered feature that exist only in ams2.
When it's a elevation change, the rear wheel drive for a moments behaves or like a gocart with no suspension or like heavy truck. What I mean you can see on the video:
Unnatural Wheelspins with smoke while it does not influence the slip behaviour, which is super weird and never seen it in real neither in other games.
Why does it happens in ams2?

Thank you.
Ps: ams2 is great, looking forward to FIA GT DLC extension

 

What I see in that video is one wheel slipping/spinning (longitudinal slip) due to the limited-slip differential. It looks to me like it's behaving just like it's supposed to? You mention gokarts, but gokarts don't have a rear differential at all, so both rear wheels always rotate at the same speed.

In the animation below, water is used to indicate loss of traction, but in your video you can substitute the water with elevation change (the wheel being raised off the ground)

 

Yes, I know it does not have, but in this moment cars feels like it , like it has no suspension and hard metal chassis. I know it's weird ) the car on the moment just slips, but don't have sharp grip loose, snap or whatever ...
Also it does not have water on the example.

But also never seen any of this obvious behaviour that includes smoke in real racing. Especially the smoke part...

So that's why IAM curious

 

Real racers generally are trying to avoid it , it’s not conducive to good tyre management.
But it’s exactly as I would expect the unweighted wheel to respond.
I do it all the time in my work vehicle on steep tight bends .
The smoke you could argue is a little bit more artistic than you might see in real life , but that can definitely depend on the amount of power the car has to rapidly spin the wheel.

 

Have you ever driven an open diff RWD car or a weakly locking rear diff car over a crest IRL? That's exactly what happens.

As you go over a crest while hard cornering, the inside rear tire gets severely unloaded and will light up the inside tire. We have a saying for that, "One tire fire".

Basically the torque is escaping out the unloaded inside rear wheel causing it to spin. While less of the torque is going to the outside rear tire that is still loaded. You may even see a puff of smoke in those conditions, but more likely not. The biggest indicator will be a slight increase in revs without forward acceleration. Those situations don't really upset the chassis with a snap of oversteer; as the outside tire which is carrying most of the load and doing most of the cornering is still within it's traction circle.
This is perfectly normal behaviour if you have less preload or clutches in your diff leading to less lockup.

 

Hi, no offense intended, but I need to comment on the R25 and R26. I've been driving these F1 cars in various simulators since 2007, and it already happened with VCR in Assetto Corsa that they made a very nice-looking R25 with a spectacular sound, but it was very unbalanced in terms of downforce and mechanical grip. It had insane downforce, higher than a 2010 F1 car, and very poor mechanical grip. There was more than a second of poor car balance; in sectors with fast corners, you were about a second faster than in reality, and in the last sector of Hungary, about a second slower. It was terrible. Now I've been watching many videos of your new R25 and R26, and it seems the balance is very good! But there's something I think is wrong, and that's the traction control. Those "skids" coming out of corners, even at over 150 km/h, are not typical of these cars. It was almost impossible for these cars to spin their wheels at high RPMs coming out of even a 50 km/h corner. In fact, I see them spinning almost as much as the R28 without traction control. It's as if the traction control is only working at 30%, and I've seen this in all the videos uploaded with these cars. Does anyone else experience this? Is it a flaw in the game engine itself, which doesn't simulate traction control well, or is it a problem with the mod? I'm just mentioning it in case it could be improved, because I really think it's a significant flaw, and it frustrates me because the rest of the car's handling in terms of cornering, braking, and inertia seems incredible. Thanks and regards.

I've attached a video from 2006. Even before you're halfway around the apex of the curve, you can already floor the accelerator pedal. The traction control does everything; there's no loss of traction, just a slight loss of "power" because the accelerator is being regulated from below.

 

Hi Reiza Team,
I wanted to share some honest but constructive feedback. I’ve actually loved the foundation of Automobilista 2, but lately—especially with the whole patch rollercoaster since the 1.6 update—I find myself feeling more and more disconnected from the actual driving experience.
I use both the PlayStation VR2 and the Meta Quest 3 on my PC, so I can say with confidence: Graphically and in VR, AMS2 is an absolute powerhouse. Visually and performance-wise, there is simply no better simulation on the market. The immersion inside the headset is an absolute masterpiece and leaves the competition far behind.
Which makes it all the more painful as a VR enthusiast when the driving physics in the limit zone completely break that immersion. It feels like Reiza is currently bouncing between two extremes and completely missing that realistic sweet spot in the middle.
When the 1.6 update first dropped, the physics were way too nervous. The moment the car started sliding even a tiny bit, it was instantly gone with zero chance of saving it, which felt extremely unnatural. In response to the complaints, the subsequent patches swung completely into the opposite extreme. Now, it’s almost impossible to actually lose the car. If I drop a wheel onto the grass in a car today or if the rear steps out in a corner, it’s way too easy to catch compared to other sims. Instead of the aerodynamics and grip snapping realistically, the whole car just goes into this highly forgiving four-wheel drift that you can save effortlessly. In the limit zones, this just doesn't feel like realism anymore—it borders on an arcade racer and takes away all the thrill of pushing a car to its edge.
I honestly think a big part of the issue is the sheer amount of content. AMS2 has an incredible number of vehicle classes from all kinds of eras. On paper, that’s amazing, but it feels like the team is stretching themselves too thin. Whenever global tire updates are pushed, certain classes suffer massively because it's impossible to fine-tune the physics and aerodynamics for every single one of them. In my opinion, it would be much more sustainable to focus on a more streamlined selection of vehicle classes and get their handling absolutely flawless, rather than constantly cramming more variety into the game. Quality and depth should really come before pure quantity here.
On top of that, the game is still missing the core pillars needed to back up this grand VR experience. To this day, multiplayer lacks a proper, officially integrated matchmaking or rating system. Relying on external website solutions just can't replace a real, built-in system, leaving online racing feeling like a total lottery. And as a single-player driver, you're still waiting in vain for the promised career mode.
Right now, driving in AMS2 feels like a compromised experience trying to please everyone—but in the end, it just alienates the players who are looking for a true, authentic challenge. The game has all the ingredients to be the absolute number one, and I really hope Reiza finds the courage to bring the physics back to a realistic middle ground and focus on the depth of the simulation.
Ultimately, I’m only sharing this because I’m a huge fan of what you've built, and I desperately want AMS2 to become the definitive sim it rightfully deserves to be.
Thanks for reading and for all your hard work!

 

You make the mistake of comparing to other sims when claiming it's too easy to catch it. The developers of any sim don't use another som the target for the physics. DO you have personal IRL racing experience of some cars in the game or at all or do you have some other material as proof? You also definitely should be specifiying some exact cars you think this would be the case and not generalizing it to somehow mean all cars in the game.

In any case, I do agree with the assesment that after 1.6 (might have been earlier on some cars actually), when you lost it, that was pretty much it and the fact that now it's actually easier to catch it, but I personally don't feel it's somehow too easy or arcade that you can (but I don't have any more proof to make my opinion any more true). There is definitely the point of no return and if you haven't found it, you probably aren't at the llimit or at most, the issues are limited to some car/cars that I haven't driven.

Here's a example of a 4 wheel slide IRL and Coach Dave talking about how far you can go:

(just check from the timestamp, the comparison to other sims is irrelevant) IMO, slides like that on AMS2 that you can catch seem realistic to me.

Also, different tires, different brands have different qualities even with same car and we know the sims are modelling different brands, with some you can get very peaky/snappy that have very small window to bring it back, some might have different qualities and allow more in that regard (but then have some other traits, like picking up marbles when hot too much as described by Coach Dave in very recent video).

I find the physics generally really nice on many cars now and definitely find it exciting to drive on the limit, which I have to do to compete against some really fast drivers on the game out there.

This is physics discussion though and that would have it's own thread..

 

Automobilsta is like a box of chocolates — you never know what you're gonna get."

A lot of content also implies, in a way, a lot of problems.

 

[/QUOTE]

No, I'm not a pro racer, but I do have real-life track day experience pushing high-horsepower sports cars to their limits. When you drive a car like that hard on a real track, you quickly learn how brutal and unforgiving the real "point of no return" is. There is no casual drifting past that line. And this issue in AMS2 isn't just limited to a few niche cars—it’s highly noticeable across the main Formula and GT classes.
Also, comparing to other sims is completely logical. Since we lack g-forces at home, cross-referencing other top sims is the only way to evaluate how different dev teams translate the same real-world data. When AMS2 lets you easily catch slides that would put you straight into the wall in iRacing, ACC, or real life, that’s a valid observation.
Your Coach Dave video actually proves my point, but you're missing a crucial distinction: in that video, the 4-wheel drift is intentional and controlled as a driving technique. In AMS2 right now, it feels like the physics engine is artificially catching an unintentional mistake after the fact. The recovery window is just way too wide.
If you enjoy the current state, that’s totally fine. But for my taste, Reiza pushed the pendulum a bit too far into safety-net territory with the recent hotfixes.

 

I'm not a pro racer either and I also have some real world race experience (auto-x, Time Trails and W2W Endurance racing). I think the cars that have V1.7 physics update feels very natural and believable.

Remember the tires you use IRL play a big part in how much slip angle they can tolerate. Some compounds tolerate more, some less. Generally, high grip slicks tend to tolerate more. This is why a lot of Pros who raced in iRacing's Indy car series during the pandemic blasted iRacing's unrealistic tire model. Joseph Newgarden said you can lean on the tires a lot more IRL and play with the slip angles as you are driving. That there was actually a pretty wide range of play there. This has been my personal experience as well.

Personally I think, comparing sim to sim is the wrong method. You should compare what a sim does to your personal experience of racing cars around a track. If you have enough real world experience, you'll immediately know what feels correct and what doesn't. Also, it's actually a pretty well known fact, you can drive cars that have added aero more aggressively on track without the rear being very unstable. My personal experience, V1.7 physics and tires feel very believable. You can play with the slip angle and dial it back or go over the limit and lose it. That being said, even people with real world experience disagree about which sims replicate realism better.

Here's a video of me setting a new track record for my TT class this year.

 

Also the pro-drivers tend to vary on what sim they prefer the most. iRacing's model is famous for forcing you to drive under the limit and some pro's actually seem to like that and some don't, some thing LMU feels the most correct, some think it's AC and so on. It might be their preference or just that the cars they race in irl and the tires they are closer to the sim what is their preference.

Now that you can better regain grip, AMS2 feels a lot more natural to me. Could be that it might be slightly too forgiving on some cars, that the window to do that is too wide, but it's definitely a lot better than it was before IMO and I have plenty of fun being push it to the limit and definitely can't catch them all when doing that.

Also, you definitely won't gain lap time when having to do that!

 

I'll keep this short,
I have often criticized Reiza, like I didn't like 1.6 physics.
and currently are still some there,
but not about physics right now!
That's really well done,
iff Reiza has planned to change something with phyisics, please, very, very carefully!!!
quickly messed up.

 

Your observation is totally valid. But people will oppose to you, because don't expect people will spend time confirming if there is anomaly here. I have spent some time looking in to this, because I also noticed this some time ago.

Those who replied pointed out that unloaded tire will slip and they are correct. Its very hard to tell, but I could guess that in AMS2 either slip happens too easy or when it does it spins up tire more, or doesn't regain traction for longer then IRL, or cars weight and downforce is not generating enough downward pressure.

Taking your example of Mount Panorama, driving over crest on mountain straight (T1 to T2) will cause slip. Again, which does happen irl, but its not as extreme. Here both tire unload equally which reduces diffs influence. And if you look at Motec data tires still remain in contact with the ground, wheels spin (by around 20 ω) and engine RPM rise by about 600 to 800 rpm. When I did look in to where I could find IRL onboards with visible engine RPM, the max rpm spike i could find over this crest was 100 RPM, but mostly only hit maybe 40 or 50 rpm spike. The same thing in cutting, there isn't any extreme rise in engine RPM like you can get in AMS2 running, what I would consider, reasonable diff values.

It is true, locking the diff will close to eliminate wheelspin in the cutting, increasing rear wing does reduce slip over mountain straight crest. But you have to go to extreme values and even then together with different rebound or spring settings I was able to get "realistic" results.

Personally I believe, there is something that, to me, is slightly off to how it drives IRL. I also think this same issue also has effect whenever you are driving over higher curbing and one corner of car gets unloaded, tire still in contact with ground and you can have these extreme RPM spikes/wheelspins. This together with TC and ABS to me are the only big downsides right now, but generally the physics and driving experience have become really nice over last couple years.

 

If it’s able to be reduced through set up changes , then that’s just a case of the set-up not being ideal for this track . ( which is quite unique)
In general anything that can be tuned out with reasonably standard changes is not a physics issue, but just a set-up issue, but there’s a lot of track combos and only one initial base set-up for a car , your never going to find the perfect one out of the box .