Physics discussion thread

Please record some footage how you get the frontengine cars to slip on corner exits with warm tires. It's bascially impossible to tell without anything to work with.

In the video I recorded about my overlays you can watch some M4 on Sebring footage of me, starting at like 9:05. I think even looking at the footage the car looks as planted as it felt when driving

 

I think that floatiness perception comes from the FFB. Because Default vs default+ vs DanielKart FFB feel quite different than each other even thou the physics is the same. I feel the tire bite more with custom FFBs. I feel more connected to the road with the latest update, especially with GT3.

 

There's never going to be a sim thats not acused of floatiness, its as simple as that..
Having said this, for me, theres a massive difference in the new GT cars (GT1, GT2 and GTR) vs the existing "enduro" GT cars in the sim (GT3s, GT4s, GTEs, etc)

My hope is that someday the GT3s can feel as good as the GT2s or GT1s

 

yeah, the F-Junior was one of my goto cars. You could steer it nicely and controll the rear with the throttle and get great turn in etc. Now it seems like an understeery 4 wheel drifting pig. Really feel like the fun has been sucked out of it.

I swapped to the formula Vee Gen 2 for my throttle steering pleasure now. Great little car this one.

 

From all the sims I have played, I'm not sure do GT3's feel amazing in any of them You still can get fun races with those and it's also easier to just jump in and drive for many, but what comes to driving itself... IMO there are a lot much more enjoyable cars to drive.

 

As someone who rides skateboards in real life, you do not want to be sliding all over the place on a skateboard. Maybe a snowboard! They can move freely in any direction . But skateboards, I mean you can do reverts and such but generally you only want to move forwards or backwards.

I pretty much ignore the floaty/slidey/whatever arguments these days because AC and ACC have long been considered the 'gold standard' for racing sim physics (rightly or wrongly) and they are both slidey as hell.

Just play the game you enjoy, whether it's AMS or something else

 

I wondered about this so took the F-Junior for a spin at Laguna Seca on a cool track and then Oulton Foster's with today's cool and light rain.

As for understeer, it has lots, so I look at the most basic set-up adjustments, first. What do we see? Front tire pressures higher than rears. Front ARB stiffer than rear. I reversed those proportions while also dropping tire pressures all around and voila, the car now drives like it should in terms of balance.

However, even factoring in that this car could be called the F-Trainer Vintage and is supposed to be tricky to drive, and has skinny little tires that I now see the approximate same size being used on electric bikes, there is something wrong IMO that could be confused for understeer.

This format car is supposed to encourage learning the classic slow-in, fast-out technique for racing. You generally never want to enter a turn too quickly or have to trail brake a rear-engine car. This one right now provides a very severe penalty if you try it.

When you enter a bit too fast, the car and its tires are not responding normally, IMO. It's as though no (or too little) traction is gained when weight shifts forward under coasting or braking. The tires just refuse to grip and the car careens forward regardless how much steering input is used. It also brakes very poorly (again, the brakes on these cars were pretty terrible by modern standards) relative to its braking performance in other scenarios where turning the wheel is not involved. I am sure there are multiple factors involved, but the end result of driving straight off the track when the car should be scrabbling for grip (sometimes regaining composure and sometimes eventually falling off the edge of the track) is missing for me.

The rain/wet track has very little effect on this--the tires seem quite plausible in the rain versus dry. To a lesser extent, the rear tires and therefore the rear end of the car are also don't seem to gain enough grip under acceleration (rear weight shift). They can let go more readily than I would expect, especially given the lowly output of these engines.

 

@Joaquim Pereira
"When parameters such as the center of gravity (mass distribution), the geometry of tire contact patches relative to the CoG, tire adhesion characteristics (tire model), and suspension variables (contact patch and CoG variance) are modified, the resulting vehicle dynamics should fundamentally change - that's what sims do."

These are rudimentary things. But if you want to realistically simulate race cars from five generations, making them clearly distinguishable from one another, this simply isn't enough.

Each vehicle has its own unique designs and technologies. Porsche's ABS system probably reacts differently than BMW's or AMG's. The same goes for traction control, differentials, clutches, motor, elektric, etc. Thousands of tiny parts that create a completely different driving experience each time.

Of course, you try to do them justice as best you can. But each car should feel significantly more distinct. But it doesn't, by a long shot. This becomes particularly noticeable in GT or Forza when one engine is supposed to render 700+ cars. They might feel different depending on their class, but they're nowhere near distinct enough from each other.

All current titles have the same problem.

I've been driving for about 35 years now, and I've never mistaken one car for another. Even if they were two roughly identical VW Golfs, they would still feel different. This multiplied by 5 generations of automotive technology...

Reiza, Kunos, Studio 397, and all those other amazing companies. They have virtually no chance of creating true reality here. It's all just a vague approximation.

The only real thing is the fun you have with these simulators.
If the fun is right, slipangel and floating wheels ect....blabla... are just empty words.

 

And I drive the car since it it is in Ams2 and I never did anything with the setup and the car drove well.after the update(s) it is no more .There is definitely something wrong with it. And that not the only one with historic or vintage tires.IMO.

 

When things get updated, sometimes the default setup may require some changes and it may be delayed. I wanted to check that out, and the set-up definitely allows one to tune the normal understeer versus oversteer balance. But I am agreeing there is something else wrong here. I am sure Reiza will look at it and get the car back to, or even better than it was previously.

 

But F junior has great feeling on dirt now

 

That would be great because it used to be an absolute blast racing this around Azure circuit at night.. YES.. at night

 

Come one …how the FFB CAN INVOLVE in slipping at the exit of the corner .
Example ( i don’t know how to take footage but….i believe I can express my opinion and anyone can ignore it..) i make an experiment.
i drove the same car McLaren 720 s Evo at Silverstone track in AMS2 and ACC.
Sunny day , default settings abs and tcs 1 ….
I am not fast driver i make 2.02.125 in ACC and 2.04.689 in ams2
At the curve between farm and village curve the pin at 5th lap at the apex i give the same (according throttle index) throttle. At ACC I feel that I have more to give at AMS2 the car make wave motion and I am loosing my track .
My set is the MOZA R9 with the KS wheel .
That is when I send slippery effect.
It can be that controlled by FFB ? If yes what are the calibration that you have make at pit house and in game ? Can you share them ?
The only thing that all want is to enjoy the driving …

 

There is no point in comparing two different games' track times. Every game has its own BoP and different physics. You should compare your time with the leaderboard in AMS2.

 

I don’t think AMS2 is at the point of accurately simulating different ABS systems. I suspect the existing one isn’t realistic by any competitive racing standard—but that’s just my guess. And perhaps it never will be; when that becomes the main issue with the simulation, I’ll probably be out of the discussion anyway (it would be demanding too much from a simulator of this kind).

You’re right about the multitude of systems that influence a car’s behavior. I did refer to that in my comment, but all of them ultimately contribute to the main forces acting on the vehicle—the object within the physics model. For instance, ABS and traction control affect the resultant tire forces, but that’s more in the realm of refinement. We can always increase the simulation’s fidelity, though hardware will always set the limit.

As it stands, however, AMS2 still fails to implement some basic aspects correctly. I’ve been pointing out that collisions are a major issue, as they don’t produce believable effects on the car. This is far worse than the shortcomings in the tire model, which is already quite good overall. The tire model could use further refinement in how grip is gained and lost, while keeping the maximum and minimum grip levels as they are—those, I think, are already quite believable.

If we try to drag a heavy box across the floor, we need x N of force to get it moving, y N to keep it moving at a constant speed, and then a decrease of z N to bring it to rest (y-z results in box stop moving). Each of these variables has a different absolute value.

If we replace the box with a tire in the current model, I’d say x is too low and z is too high—we should need more force to get the tire to start slipping significantly, but less force for it to regain grip and return to an almost full-grip state. Some call this tire “bite.”

This hysteresis in friction is something I find missing in the current model.

 

What initially IMPRESSED me with the AMS2 physics is how distinct each car drives even in the same class. The base characterics of each GT3 car are vastly different, but comparable between most sim titles.

Then there’s of course the base handling of each sim and ofc that doesn’t change between the cars.

What i’m trying to say is if i pick for example the AMG GT3 in iRacing, ACC or AMS2 the characterics of the car is very similar between the titles. The physic engines are just different and the engine determites ”basic flavor” of each sim.

 

TC and ABS at 1? Why so low on AMS2? Even commercial "esport" setups often run ABS at 6-8 and TC 3-5 (only qualifying setups run lower). If you run low TC, of course it will require more sensitive manual control with your throttle.

AMS2 has more correct throttle behavior than ACC and sims from Kunos. I dont' think even ACE still has it correctly implemented. So it's not surprising that you get different behavior there in different sims because of that alone (I assume you aren't 100% throttle here). This is just a known fact, not my personal critique towards ACC.

Both can have also the differential under the hood or in the setup set differently, with more locking on AMS2 on throttle than on ACC.

Track's grip levels can be different in different sims and so on.

I find it's interesting that people very often mention this slidey feeling on throttle as something that bothers them... to me it's totally opposite, having some on throttle rotation and/or wheelspin is something that makes the car feel powerful for me (I get this "f... yeah" feeling). The TC also in AMS2 doesn't kill the on-throttle rotation totally, so maybe it's one of the reason why it feels more slidy than some other sim.

 

i am hearing what you say i would like to add here that i have a throttle dead zone at 8% to avoid such impact but...........still
at real racing the TC is made to avoid slipping TRACTION CONTROL thought so even at 8 when i met that feeling i find that has some how to kill that throttle effect
my opinion

 

The whole throttle curve is different (from what I understand).

But yeah, the TC doesn't kill the on throttle rotation, you can try reducing the diff clutches to 4 or even 2 to see if it's more to your liking (probably will hurt the potential lap time though, but depends on track).

 

I cannot say if you change your setup or not, but for me Sebring requires a bit more extreme setup changes due to the bumpiness nature of the track. Default suspension settings for me always feels a bit too soft on other « regular « circuits, but in Sebring I find that softening suspension further and softening ARBs and increasing a bit the height of the car helps me on having more predictable consistency when driving around Sebring and in braking as well.