Automobilista 2 V1.2.3.0 Update Released on Steam

Automobilista 2 August Update 01.jpg
Reiza Studios has updated Automoblista 2 to version 1.2.3.0, and added two new cars, improved Real Weather features and another round of fixes and improvements.

Automobilista 2’s August 2021 update is now live on Steam. In typical fashion, developer Reiza Studios has offered new, free content plus several other improvements and fixes to the racing sim.

The new content this month is a classic American muscle car in two flavours. The Chevrolet Corvette C3 has been added to both the Vintage Touring Car class as well as to the GT Classics Class in the "R" Spec to compete with the recently added Porsche RSR. The Corvette C3 was teased last month but was offered only to AMS2 beta users before this build.

Another significant improvement that has been added in this build is the addition of accurate environment temperatures in the Real Weather system. The Real Weather system in AMS2 allows users to select accurate current weather for track locations, and even historically accurate weather when past dates are chosen.

On the physics front, many of the cars have been given a tire tread adjustment to go with the potentially significant changes to the clutch inertia and tire tread fixes.

Automobilista 2 August Update 03.jpg


V1.2.3.0 CHANGELOG

CONTENT
  • Added Corvette C3 to Vintage Touring Car Class
  • Added Corvette C3 "R" Spec to GT Classics Class
GENERAL
  • Added Damage Scale option
  • Added option to allow/disable pit stop refuelling
  • Added time +1 lap option for race sessions
  • Real Weather now uses real environment temperatures for all tracks (including historical data)
  • Added weather data between original feature release on June 4th through today to historical database of all locations
UI&HUD
  • Lobby Page session details: Fixed incorrect value for mandatory stop; Removed duplicate track cut entry; Fixed rolling start label; Added formation lap info
  • Fixed Lobby driver list scrollbar
  • Fixed Start button still being available when a championship is complete
  • Update various vehicle class colours
  • Corrected track details for Azure
  • Fixed missing track map for Interlagos Stock Car variant
  • Fixed Start button still being available when a championship is complete
  • Further restricted setup options showing redundant adjustable settings in fixed configurations
PHYSICS
  • Fixed bug in intake manifold pressure model causing poor throttle response in several cars
  • Minor tire tread adjustments to all GT classes, Stock Cars,, P2, P3, F-3, F-Trainer, Procar, Group A, Group C, Porsche RSR, F-Vintage, F-Vee, Opala 79, Old Stock, Lotus 23, Street Cars
  • Revised clutch inertia values for all cars
  • Porsche RSR 1974: Tire carcass & tread model adjustments; slightly adjusted CoG height & inertia; Minor default setup adjustments; Fixed wet tires overperformance
  • Fixed redundant 1 bumpstop level adjustment on setup screens for several cars that don´t have that setting
  • F-Vintage: Added extra final drive ratios to both V8 models as well as Lotus 49C, Brabham BT26
AI
  • Altered AI selection logic to selected fastest drivers first when fixed drivers are unavailable or exhausted.
  • Fixed excessive clutch inertia causing AI slow starts & gear shifts for Procar, Group A & Porsche RSR
  • General wet weather callibration pass for all cars
  • BMW M1 Procar, Group A, Porsche RSR AI callibration
  • Improved line and AI performance at Interlagos, Montreal & Santa Cruz
AUDIO
  • Added diversified curb, grasscrete, astroturf sound effects according to material type & profiles (all cars)
  • Added dirt pickup sound effect (all cars)
  • Corrected various curb material sounds for Granja Viana, Ortona, Long Beach, Silverstone 1975, Silverstone 2001, Spa-Francorchamps, Kyalami, Imola, Kansai & Montreal
  • Adjusted audio pan compensation for curbs and added support for more types to road noise sound
  • Mercedes 190E DTM: Fixed no engine sound when looking back in chase cam view
TRACKS
  • Minor adjustment to visual Livetrack groove
  • Hockenheim: Fixed minor object popping issues
  • Salvador: Fortified some gaps in the walls; Added more tree trunks to open areas; switched on collisions for stactic vehicles
  • Silverstone: Added a missing curb on old Abbey-Bridge road; Fixed a minor selection set error
  • Silverstone 2001: Reduced road noise on GP layout; Minor art & optimization pass
  • Brasilia Outer: Fixed minor clipping issues
  • Granja Viana: Improved track cut limits; Minor art & optimization pass
  • Silverstone 1975: Minor art & optimization pass; Fixed pit wall collision
  • Long Beach: Fix broken transform on some fence speakers; Adjusted pitlane cheat block
  • Interlagos: Minor art & optimization pass
  • Cadwell Park: Minor art & optimization pass
  • Adelaide Historic: Added pitlane cones; Replace triggers with new boxes; Minor art & optimization pass; Slightly reduced road noise
  • Spa Francorchamps 2020: Minor art & optimization pass;
  • Kyalami 2020: Minor art & optimization pass
  • Imola 2021: Minor art & optimization pass
  • Montreal 2021: Minor performance & art pass; Fixed strange collisions at T14 exit.
  • Added VR cams for Brasilia Outer and Curvelo B
  • Fixed missing pitbox marker in race session for Daytona and Long Beach
  • Santa Cruz: Add cones to mark pit ent/exit
  • Kyalami Historic: Revised physical terrain material types; Minor performance pass
  • Minor fixes & adjustments to trackside cameras of Kyalami, Azure, Cascais
VEHICLES
  • Metalmoro MRX (all variants): Added driver animations; Detached the cockpit gear shifter. Adjusted the cockpit view to match the new driver position.
  • F-Trainer: Add rear chassis parts and rear suspension for cockpit view
  • Stock Corolla 20/21: Fixed LOD C glitch
  • Various minor art improvements to Vintage GT helmet & suits
Automobilista 2 August Update 02.jpg


Be sure to share your thoughts on the new build below in the comments, and stay tuned to RaceDepartment for our interview with Renato Simioni of Reiza Studios coming very soon.
About author
Mike Smith
I have been obsessed with sim racing and racing games since the 1980's. My first taste of live auto racing was in 1988, and I couldn't get enough ever since. Lead writer for RaceDepartment, and owner of SimRacing604 and its YouTube channel. Favourite sims include Assetto Corsa Competizione, Assetto Corsa, rFactor 2, Automobilista 2, DiRT Rally 2 - On Twitter as @simracing604

Comments

Well it will depends on your perception of what makes a good sim. Obviously each one of them has their goods and bads. I suggest you watch some reviews on YT or make a summary from this forum and compare it with your own experience in AC.
For me, AMS2 tires feel more alive than AC, but AC support mods. AC has reached its final state but Content Manager makes it exciting again. AMS2 still being actively developed and there are a hits and misses on each build update. AMS2 has fantastic weather system and many people praise its VR capability. etc etc.
AC is like Mrs. Merkel, the end of a non-vision era. Let her go, let it go …. All those tries to polish AC reminds me on GTR1. They all have had there good times
 
Welcome to reality, LOL! An unmodified C3 is a boat by modern standards. It is softly sprung and has mushy power steering & brakes.
I agree, it's a mushy boat lol

Wouldn't reality still have forward weight shift under braking especially with with softly sprung suspension?....which is what I'm lacking in feel with many cars in AMS2 atm.
 
Last edited:
I guess it will be the same day that rFactor 2 fixes its weird momentum physics:


Or the day AC fixes its weird momentum physics:


Or the day R3E fixes its weird momentum physics:


In fact every racing sim has weird momentum behaviour because tyre models lose all relevance once the car loses contact with the road and there's little benefit to spending CPU cycles on modelling soft-body physics in a racing sim.
don't mind.
This troll took a sequence out of a YT, were idiotic drivers fail all over different sims.
As far as i can read, the sequence points to AMS1 not AMS2 :roflmao:
really good example.
i would say: FAIL of the troll :laugh:
 
Last edited:
I agree, it's a mushy boat lol

Reality would still have forward weight shift under braking especially with with softly sprung suspension....which is what Im lacking in feel with many cars in AMS2 atm.
Spring stiffness doesn't have any effect on load transfer, before we start going really off track. It's not gonna do anything to longitudinal, and only the distribution between front and rear will determine the ratio of lateral. Soft cars don't transfer more, and stiff cars don't transfer less.

Steering geometry and SAT curve will largely determine how a car feels and generally steering column torque will decrease in manual rack cars due to self aligning torque before the front is at peak force. It doesn't mean the fronts are over the limit, or even at it.
 
I couldn't agree less. In dry it's way better then AC. Compared to ACC: it's not better also not worse but it's different. Both have advanced physics. I don't race in rain so I couldn't say about raining conditions. The VR experience is not comparable, AMS2 is way ahead of AC and ACC regarding VR.

In my opinion AMS2 is the best sim atm, BY FAR. The content is all of high level, the tracks, the cars are almost all great. The graphics are the best also, the sound is okay(ACC is a bit better on this subject imo but they are improving the sounds with almost each update), the physics are since update 1.2.2.0 awesome. Opinions are different here, many are fan of AC, but in my opinion AC is outdated, the game and the interface is outdated, especially the tracks look terrible compared to AMS2, the physics are also less advanced then AMS2 and ACC. Especially in VR the experience is the best in AMS2.

So yes in my opinion it IS a good sim, a GREAT sim in fact, a serious upgrade on AC. The only thing that it lacks at this moment is enough people in multiplayer, sometimes it's great, sometimes not enough people.
I don't find VR an issue at all in AC, performance is actually better for me than AMS2 at night.
I think the best thing about ams2 is its easy to setup & get racing with AI. I love AC & content manager with shader mods, both games are good VR racers.
 
The only real main difference between empiric and physical models is how the input is generated.
Isn't it true that physical models can more easily be tied to tire animation for tread & sidewall flex?
 
Premium
LOL complet controls are gone after the update :laugh: . So my question is - have we again to delete the documents folder:D?
 
Isn't it true that physical models can more easily be tied to tire animation for tread & sidewall flex?
When comparing something like AC where the sidewall "doesn't exist" to SETA where it seemingly exists at least in the inputs? Yeah, maybe. There will likely be parameters for more things that you could get clues for the animation from.

It's not really a trait of physical models though, it's a side-effect of *needing* to have those parameters to produce the contact patch behavior. An empiric model can completely skip that if they just want accurate contact patch behavior. Remember that (ideally) the sidewall isn't in contact with the ground and doesn't actually do anything when it comes to slip that you can't do "directly" via the contact patch.

What is officially said about SETA is that "The carcass simulation used in Project Cars is a finite element simulation with specific computational optimizations specific to real time tire simulation" ie: they likely use a physical based model with parameters for the construction of the tire and the properties of the contact patch to produce pre-generated parameters/curves, and those parameters/curves are then used in the brush model to produce the real-time interactive behavior.

At that point there should not really be any fundamental difference between say AC and SETA apart from that getting accurate parameters is unlikely with purely the physical generation process. An empiric brush model like AC's isn't necessarily "less complex" in what the tire *actually* does when you run it in the simulation in real-time.

If CSP decides to add a "sidewall" simulation to AC to get more variable and interactive pressure effects for example (buckling, debeading etc. at low pressure is one example) then you would have similar inputs to do animations from. The difference would be that how your curves respond to low pressures and any additional effects from them would not be directly 1:1 tied to for example what the slip peak of your tire is or how stiff vertically it is. In an empiric model you can specify them separately.

Physical models are not "tire tools" by the way, they still have a bunch of esoteric inputs and hard-to-determine physical parameter inputs that you'll only really find out via testing or correlation. The big difference IMO is that you'll always get a better result with less testing/correlation with an empiric model because you can go over it one-by-one with little to no risk of becoming volatile as long as you know what you're doing.

The downside is you need to either use some money to test the tire, or a long time to correlate it; but physical models cost several millions to develop so I think in the end it'll be a bit more efficient considering that the accuracy empiric models can produce is still better. Even if you have load/slip/heat/pressure/stiffness interaction data for the tire, it might be impossible to reproduce an acceptable representation if your physical model is not cut out for it.

I believe there are probably worse and better physical models out there, I have seen some very reasonable curves out of finite-element simulations for tires, but those are not exactly game applications. Niels and many others have also developed "tire tools" that act similar to physical models, where you input some basic parameters and get out multiple tires. Although the difference in those usually is that A: They have hardpoints from empiric data and B: Have few inputs and are not super volatile. It is not so much a simulation like SETA is as much as it is a repository IMO.
 
The VR experience is not comparable
7067136696D82D0F24F247A92A7F1E46788FF79D

not comparable, but can look a little painful. :confused:
That wheel going trough the leg of the driver, is troubling in VR.
At least he is lucky, the driver in the "street C3, has the wheel going trough both legs.
Some might think it is a little disappointing, that in a brand new car release, Reiza has not figured yet how to position the driver in a car, let along not rectified all of the wrong positioned driver in countless cars since release.
Well, it is what it is!
 
Last edited:
If this Article is about AMS2, why it's getting spammed with walls of text about AC1 ?...

Its not wonder that participation on RD has been declining quickly, and i am not even blaming the mods, they have a hard time already...

But there is no joy in even reading anything here anymore when it starts resembling some random youtube comments section.
 
If this Article is about AMS2, why it's getting spammed with walls of text about AC1 ?...

Its not wonder that participation on RD has been declining quickly, and i am not even blaming the mods, they have a hard time already...

But there is no joy in even reading anything here anymore when it starts resembling some random youtube comments section.
The wall of text is about the physics of AMS2 specifically the tyre model in comparison to AC. Ironically just about the most interesting thing I've read in here in a long time.
 
When do you have started AMS2 the last time? With introduction of multiple controls profiles it was required to remap the controls.
I also ran into all control assignments disappearing with this update, but for the exact reason you mention. I had not done any gaming at all for over month due to “life reasons.”

I did want to mention, however, that the only thing related to this I recall reading about with the prior update(s) was pedals needing to be reassigned. I saw multiple mentions of that. I don’t recall ever seeing anyone mention in prior comments that they had to reassign all control mappings, just pedals. Maybe I just missed it though.
 
When comparing something like AC where the sidewall "doesn't exist" to SETA where it seemingly exists at least in the inputs? Yeah, maybe. There will likely be parameters for more things that you could get clues for the animation from.

It's not really a trait of physical models though, it's a side-effect of *needing* to have those parameters to produce the contact patch behavior. An empiric model can completely skip that if they just want accurate contact patch behavior. Remember that (ideally) the sidewall isn't in contact with the ground and doesn't actually do anything when it comes to slip that you can't do "directly" via the contact patch.

What is officially said about SETA is that "The carcass simulation used in Project Cars is a finite element simulation with specific computational optimizations specific to real time tire simulation" ie: they likely use a physical based model with parameters for the construction of the tire and the properties of the contact patch to produce pre-generated parameters/curves, and those parameters/curves are then used in the brush model to produce the real-time interactive behavior.

At that point there should not really be any fundamental difference between say AC and SETA apart from that getting accurate parameters is unlikely with purely the physical generation process. An empiric brush model like AC's isn't necessarily "less complex" in what the tire *actually* does when you run it in the simulation in real-time.

If CSP decides to add a "sidewall" simulation to AC to get more variable and interactive pressure effects for example (buckling, debeading etc. at low pressure is one example) then you would have similar inputs to do animations from. The difference would be that how your curves respond to low pressures and any additional effects from them would not be directly 1:1 tied to for example what the slip peak of your tire is or how stiff vertically it is. In an empiric model you can specify them separately.

Physical models are not "tire tools" by the way, they still have a bunch of esoteric inputs and hard-to-determine physical parameter inputs that you'll only really find out via testing or correlation. The big difference IMO is that you'll always get a better result with less testing/correlation with an empiric model because you can go over it one-by-one with little to no risk of becoming volatile as long as you know what you're doing.

The downside is you need to either use some money to test the tire, or a long time to correlate it; but physical models cost several millions to develop so I think in the end it'll be a bit more efficient considering that the accuracy empiric models can produce is still better. Even if you have load/slip/heat/pressure/stiffness interaction data for the tire, it might be impossible to reproduce an acceptable representation if your physical model is not cut out for it.

I believe there are probably worse and better physical models out there, I have seen some very reasonable curves out of finite-element simulations for tires, but those are not exactly game applications. Niels and many others have also developed "tire tools" that act similar to physical models, where you input some basic parameters and get out multiple tires. Although the difference in those usually is that A: They have hardpoints from empiric data and B: Have few inputs and are not super volatile. It is not so much a simulation like SETA is as much as it is a repository IMO.
Big wall of conversation but you are still missing the point: the SETA model allows to account for the tire flexing in real time which cannot be accounted for in real time with the model AC and ACC use. That is why their tires feel like bricks and AMS2 tires feel alive.
Since you seem a genius of engineering, you should be well aware that tire wall flexing has a deep impact on the forces exchanged at the contact patch: the easiest way to see that is that F1 cars next year will have a very different range of suspensions tuning compared to this year because the tire wall stiffness will be completely different. A model that is not capable to account for that flexing is a model that is missing pieces of information like the correct vertical and lateral loads at the contact patch and all the forces that come off of them including FFB.
This is the reason why in ACC you have a fixed optimal pressure: the tire model has significant limits that the SETA does not have.
AMS2 cars can occasionally be out of tune in their complex parameters, but AC/ACC with their models are out of opportunity to give a proper tire feeling for the rest of their shelf life.
 
Big wall of conversation but you are still missing the point: the SETA model allows to account for the tire flexing in real time which cannot be accounted for in real time with the model AC and ACC use. That is why their tires feel like bricks and AMS2 tires feel alive.
Since you seem a genius of engineering, you should be well aware that tire wall flexing has a deep impact on the forces exchanged at the contact patch: the easiest way to see that is that F1 cars next year will have a very different range of suspensions tuning compared to this year because the tire wall stiffness will be completely different. A model that is not capable to account for that flexing is a model that is missing pieces of information like the correct vertical and lateral loads at the contact patch and all the forces that come off of them including FFB.
This is the reason why in ACC you have a fixed optimal pressure: the tire model has significant limits that the SETA does not have.
AMS2 cars can occasionally be out of tune in their complex parameters, but AC/ACC with their models are out of opportunity to give a proper tire feeling for the rest of their shelf life.
If you'd like to lay out a mathematical definition for why you need a separate sidewall flex model to achieve correct FY and FZ at the contact patch, be my guest. It sounds to me like you haven't actually done any sim dev but you have your chance to educate me.
 

Latest News

Article information

Author
Mike Smith
Article read time
4 min read
Views
24,348
Comments
150
Last update

Online or Offline racing?

  • 100% online racing

    Votes: 74 7.3%
  • 75% online 25% offline

    Votes: 106 10.4%
  • 50% online 50% offline

    Votes: 147 14.5%
  • 25% online 75% offline

    Votes: 279 27.4%
  • 100% offline racing

    Votes: 407 40.0%
  • Something else, explain in comment

    Votes: 4 0.4%
Back
Top