Part One of Automobilista 2’s November 2021 Development Roadmap Revealed

Article Discussion
Automobilista 2 November Update 01.jpg
Reiza Studios has released the first batch of information on updates that Automobilista 2 players can expect in the month of November.

The November roadmap for Automobilista 2 is so large that Reiza Studios has opted to reveal it in two parts. The first part was published as a Steam article. Here are the highlights:

Players can expect two new Volkswagen cars – We published an article recently about the addition of two Volkswagen cars as part of the VW TSI Cup E-Sports series. More information is now available. The cars will be utilized in a similar manner to the Acelerados show in Brazil, which features presenter Rubens Barrichello or other pro drivers chucking cars around Velo Città in a manner similar to Top Gear.

Improved differential model – Reiza has made great strides in the development of the driving physics in AMS2 since release. They were challenged with trying to recreate what they achieved in AMS1 using the Madness Engine. One aspect of the physics Reiza has been focused on lately is the differential modeling. The next major build will incorporate these improvements.

Automobilista 2 November Update 02.jpg


AI improvements and modding – The huge number of tracks and car classes on offer in Automobilista 2 has proven to be a challenge to Reiza in terms of building competent AI to race against in solo play. November will be improvements on this front, and even allow players to create their own. The initial roll out will be text edits, but this process should be polished in the future.

Racin’ USA part 2 and laser-scanned Azure – Content is always a primary focus of updates, and Reiza has offered up some big news on that front. Classic Indycar is coming, along with new US tracks. The picture shared by Reiza appears to show a late 90’s CART car. Azure (Monaco) will be overhauled using laser scan data, replacing the version inherited from Project CARS 2.

Click this link to read the part 1 of the November Automobilista 2 roadmap from Reiza. We will share more information on the November update once part 2 of the roadmap is revealed. Let us know in the comments below what you’re interested in trying out from this busy development month.
  • Like
Reactions: fragged, Alex72, StationaryRacer and 21 others
About author
Mike Smith
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

More in Automobilista 2

Comments

Stephen O'Sullivan
Peter Kerényi said:
I did report it a couple of months ago, but thanks.
Click to expand...
Ah, sorry, I wasn't to know that from your post. All I can suggest is that you try again.

The surface change on the main straight at Curitiba might have something to do with it. There have been occasional problems with different road surfaces in pit areas that have affected grip, etc. I thought Reiza had addressed the start/finish line problem but maybe that bug persists.

I'm not sure what is causing your F3/Estoril problem. That last corner eats tyres for breakfast so it might be that the AI is experiencing low grip en masse or the AI lines need tweaking.
 
Last edited:
Peter Kerényi
CrimsonEminence said:
The S/F line braking on this track was already fixed some months ago. Does it still happen?

When do you have tested it last time, just to have a rough estimate?
Click to expand...
Rough estimate, I haven't really played it for more than a month. Moving, etc. I do know I reported it around early this year, then tested it late summer/early autumn and it was the same. If it's been fixed, sorry for spreading false information.
 
CrimsonEminence
Peter Kerényi said:
Rough estimate, I haven't really played it for more than a month. Moving, etc. I do know I reported it around early this year, then tested it late summer/early autumn and it was the same. If it's been fixed, sorry for spreading false information.
Click to expand...
Well at least worth another look i would say. It should've been a track and not class specific issue and was also acknowledged, but who knows. :)
 
Boby Kim
What a battle for Reiza. Its so sad to see such talent being wasted with the aweful Madness Engine. Whatever they do, however they do it...it is Pcars and it always be Pcars. AMS2 will never ever reach the top level of AMS1
 
David Peres
Boby Kim said:
What a battle for Reiza. Its so sad to see such talent being wasted with the aweful Madness Engine. Whatever they do, however they do it...it is Pcars and it always be Pcars. AMS2 will never ever reach the top level of AMS1
Click to expand...
Please enlighten us with your knowledge of how the Madness Engine works and why it can't be tweaked to provide good results. Don't want to hear anything about PCars... I want proper data, formulas and graphs, expaining about how the engine is fundamentaly flawed in such a way that cannot be changed. I'm sure you can do that.

EDIT: Altough I think the kind of thinking that Boby is showing on his post is pure bullshit, I do have to say that I find the GT3 cars in AMS2 absolutely awfull and I like them less and less the more I try to play around with setups to make them better. Same thing for some of the modern F1 cars.

On the other hand there are many other cars that are amazing, I particularly like some of the historical stuff like the BMW 2002, the Porsche RSR 74 or the M1 Pro Car.
 
Last edited:
T
Peter Kerényi said:
Another 98% offine simmer here. AMS2 AI is the most exciting currently, when it works. If they manage to fix stuff that doesn't work (like certain combos, e.g. F3 @ Estoril just all crash in the final corner after a couple of laps, SC @ Curitiba brake at the S/F line, inconsistent difficulty, top speeds of the AI being low overall, etc.) it will become the best. ACC AI is less exciting to race but much more consistent (scope of the sim helps them).

Like somebody said about F1 engines, it's easier to develop a fast reliable engine from a fast unreliable engine than a slow reliable engine. AMS2 AI is "fast" and "unreliable" currently. The fact that they acknowledged some of this AI stuff in the roadmap and work towards improving as well as making awesome features like the AI customizer gives me that hope and excitement for the future of the game. Wondering when does another studio with inconsistent and often broken AI *ahemmmmrF2mm397mm* make this step?
Click to expand...
Fun fact is, that you can do all of the AI tweaking stuff mentioned in this roadmap in rF2 allready, but at the end nobody really uses it as people don't like to fiddle around with text files and I have the feeling that noone will do that in AMS2 either besides a few die hard fans. People just want to have some fun out of the box I guess.

That said, their approach to prioritizing development focus and their content selection is remarkable and I am mostly just peeking over in shame from my rF2 corner to have an envious look. Considering that S397 had 80s and 90s Indycar lisences at hand and all the other great cars that could and should have found their way into rF2 at one point makes you wonder what happened along the way. Luckily we still have the modding community that is filling the void to some extent.
 
Peter Kerényi
tr1v1um said:
Fun fact is, that you can do all of the AI tweaking stuff mentioned in this roadmap in rF2 allready, but at the end nobody really uses it as people don't like to fiddle around with text files and I have the feeling that noone will do that in AMS2 either besides a few die hard fans. People just want to have some fun out of the box I guess.

That said, their approach to prioritizing development focus and their content selection is remarkable and I am mostly just peeking over in shame from my rF2 corner to have an envious look. Considering that S397 had 80s and 90s Indycar lisences at hand and all the other great cars that could and should have found their way into rF2 at one point makes you wonder what happened along the way. Luckily we still have the modding community that is filling the void to some extent.
Click to expand...
Customizing AI has little meaning when the AI itself is... that incapable and/or inconsistent. I know I hold a grudge against rF2 but I can't help it. In rF2 one can have awesome AI battles and racing... with certain cars at certain tracks in single class. Which cars and tracks? Someone should make a spreadsheet I guess.
 
Kyuubeey
David Peres said:
Please enlighten us with your knowledge of how the Madness Engine works and why it can't be tweaked to provide good results. Don't want to hear anything about PCars... I want proper data, formulas and graphs, expaining about how the engine is fundamentaly flawed in such a way that cannot be changed. I'm sure you can do that.

EDIT: Altough I think the kind of thinking that Boby is showing on his post is pure bullshit, I do have to say that I find the GT3 cars in AMS2 absolutely awfull and I like them less and less the more I try to play around with setups to make them better. Same thing for some of the modern F1 cars.

On the other hand there are many other cars that are amazing, I particularly like some of the historical stuff like the BMW 2002, the Porsche RSR 74 or the M1 Pro Car.
Click to expand...
Well, from a physics dev standpoint, the SETA model being a more physical based model probably doesn't have as much capability to change specific tire behavior as more empirical based ones do.

For example if the tire stiffness is good vertically and the tire responds correctly to pressure and camber, it might be far too stiff laterally and have a too low lateral slip peak. If you fix the lateral slip peak, there is no guarantee your longitudinal peak slipratio will remain correct if it even was correct to begin with, and the vertical stiffness, pressure and camber sensitivity etc. might all fly out of whack. I suspect the thermal modeling is also linked to the tread and carcass.

In a more empirical model you can just generally input in the right parameters largely independently. In a fully physical one, you're stuck with what the model provides. It'll also depend on having difficult/'impossible' to acquire information about the tire that a consumer simulation game developer definitely does not have and will probably never have.
 
David Peres
Kyuubeey said:
Well, from a physics dev standpoint, the SETA model being a more physical based model probably doesn't have as much capability to change specific tire behavior as more empirical based ones do.
Click to expand...
That seems to be true that physical tire models make it more complicated to achieve the desired result becouse there are many variables and they affect each other in real time, which makes things more unpredictable. That's the reason why Niels prefers simple empirical models.

However, we do have many examples of cars with great tires in sims with physical tire models (including cars in AMS2) so I don't see that as an argument as to why Reiza will not be able to achieve very good results as they learn more and more about how the SETA model works and how to take the best out of it.
 
Last edited:
T
Peter Kerényi said:
Customizing AI has little meaning when the AI itself is... that incapable and/or inconsistent. I know I hold a grudge against rF2 but I can't help it. In rF2 one can have awesome AI battles and racing... with certain cars at certain tracks in single class. Which cars and tracks? Someone should make a spreadsheet I guess.
Click to expand...
I could say the same thing about any sim racing product that is out there. AMS2 and rF2 are pretty similar in that regard actually and you have some combinations that work good and some that work bad. That's just the nature of the whole system of tuning AI specific to tracks and cars. I mean, there are systematical issues aswell but in theory you can tune the AI in rF2 to a decent level from my experience. I have incrased the SC threshhold so that I don't get FCY and I know how to get proper wet weather racing aswell. I am not saying that it doesn't need more work but I have it at a level that is as good or better than other products out there. It's just comes down to how the content is setup, wich is a problem in many other sims aswell. And the bigger and more complex the products become the harder it get's to fix all the issues.
 
Kyuubeey
David Peres said:
That seems to be true that physical tire models make it more complicated to achieve the desired result becouse there are many variables and they affect each other in real time, which makes things more unpredictable. That's the reason why Niels prefers simple empirical models.

However, we do have many examples of cars with great tires in sims with physical tire models (including cars in AMS2) so I don't see that as an argument as to why Reiza will not be able to achieve very good results as they learn more and more about how the SETA model works and how to take the best out of it.
Click to expand...
Sure, but my point kinda is that if you take those cars with great tires and actually correlate it to some telemetry or tire data from a similar tire, it could very well be tremendously errorenous and unfixable without a ton of expensive work. Weird handling quirks of sims like iR, rF2, AMS2 can be seen with the eye, you don't really even need a data based approach.

Renato said something like "people just don't like it because it's different" but I dunno, just my 0.02 as a sim physics modder. I don't think it's entirely baseless.

Kind of offtopic, but from my experimentation with ISImotor's old empiric model in AMS1, the switch to SETA does not seem so crazy. It might be true they've gotten better results with it than the old one.

Personally I'm having tremendous issues getting even minimum acceptable transient/combined behavior in the old model, and it's a generally reasonable empiric model. I'm not yet convinced I'm inputting in everything correctly, but the car basically correlates almost fully to Assetto Corsa in steady state, so it seems to be an issue of the more model-limited things like how it handles combined grip or transient situations.

I did swap in other tires (including Niels') just to make sure I'm inputting parameters correctly, but no joy. Ironically 10 years ago I didn't find anything wrong with the almost same tire model in rFactor, but I guess it kind of illustrates how you can have more issues with a model if you have more specific expectations.
 
Peter Kerényi
tr1v1um said:
I could say the same thing about any sim racing product that is out there. AMS2 and rF2 are pretty similar in that regard actually and you have some combinations that work good and some that work bad. That's just the nature of the whole system of tuning AI specific to tracks and cars. I mean, there are systematical issues aswell but in theory you can tune the AI in rF2 to a decent level from my experience. I have incrased the SC threshhold so that I don't get FCY and I know how to get proper wet weather racing aswell. I am not saying that it doesn't need more work but I have it at a level that is as good or better than other products out there. It's just comes down to how the content is setup, wich is a problem in many other sims aswell. And the bigger and more complex the products become the harder it get's to fix all the issues.
Click to expand...
The interesting content in rF2 is Endurance. Endruance tracks, endurance cars. However, multiclass AI is terrible. LMPs spinning when overtaking or getting stuck behind, pitting AI not overtaking the whole lap even on tracks like Le Mans, etc.

Unfortunately no sim has perfect AI. All sims require some tinkering, and even that doesn't guarantee a flawless experience. For me, rF2 was just too much to manage, providing more frustration than enjoyment. If it works for you, I am happy for you.
 
K
David Peres said:
Please enlighten us with your knowledge of how the Madness Engine works and why it can't be tweaked to provide good results. Don't want to hear anything about PCars... I want proper data, formulas and graphs, expaining about how the engine is fundamentaly flawed in such a way that cannot be changed. I'm sure you can do that.

EDIT: Altough I think the kind of thinking that Boby is showing on his post is pure bullshit, I do have to say that I find the GT3 cars in AMS2 absolutely awfull and I like them less and less the more I try to play around with setups to make them better. Same thing for some of the modern F1 cars.

On the other hand there are many other cars that are amazing, I particularly like some of the historical stuff like the BMW 2002, the Porsche RSR 74 or the M1 Pro Car.
Click to expand...
I agree with you regarding the gt3 cars. They just don’t feel right. Same for the Carrera cup cars.

each sim has its strengths and weaknesses. In Raceroom, it is the open-wheel cars that I just don’t care for.

In AMS2, I always end up driving the Brazilian Stocks and the historical F1s. The Brazilian Prototypes are also a blast.
 
T
Peter Kerényi said:
The interesting content in rF2 is Endurance. Endruance tracks, endurance cars. However, multiclass AI is terrible. LMPs spinning when overtaking or getting stuck behind, pitting AI not overtaking the whole lap even on tracks like Le Mans, etc.

Unfortunately no sim has perfect AI. All sims require some tinkering, and even that doesn't guarantee a flawless experience. For me, rF2 was just too much to manage, providing more frustration than enjoyment. If it works for you, I am happy for you.
Click to expand...

Same exact issues I had. I gave up on rF2 over a year ago because multiclass racing against the AI was impossible. Over half the field would be beat to hell or retired due to constant crashing between classes, AI spinning out for no reason at all in the same spot on certain tracks like clock work. The pit logic would be completely off due to fixing damage or just pitting for some other reason. The blue flag syndrome was not great. The potential was there, but S397 progress is beyond slow.

Meanwhile AMS2 AI still need work, but the foundation is at least there and I'm hoping with the AI customization coming we will be able to dial things in just that little bit better.

I'll also say that there is a lot of talk about how "slow" developement of AMS2 is and while updates to the player base may seem like they take forever, the progress AMS2 has made since release to where it is now in comparison to other sims that have been out much longer I think the idea that Reiza is "slow" needs to be retired. In comparison to their peers I'd say they've made quite a bit of progress and are delivering on what their stated goals are in a faster manner than other studios.
 
Last edited:
G
AMAZING updates once again !!! Really looking forward to this.

I hope that we will see Sebring and Zandvoort 2021 one day in AMS2.. but all other content is so good already so no complaints :D
 
K
Tar Heel said:
Same exact issues I had. I gave up on rF2 over a year ago because multiclass racing against the AI was impossible. Over half the field would be beat to hell or retired due to constant crashing between classes, AI spinning out for no reason at all in the same spot on certain tracks like clock work. The pit logic would be completely off due to fixing damage or just pitting for some other reason. The blue flag syndrome was not great. The potential was there, but S397 progress is beyond slow.

Meanwhile AMS2 AI still need work, but the foundation is at least there and I'm hoping with the AI customization coming we will be able to dial things in just that little bit better.

I'll also say that there is a lot of talk about how "slow" developement of AMS2 is and while updates to the player base may seem like they take forever, the progress AMS2 has made since release to where it is now in comparison to other sims that have been out much longer I think the idea that Reiza is "slow" needs to be retired. In comparison to their peers I'd say they've made quite a bit of progress and are delivering on what their stated goals are in a faster manner than other studios.
Click to expand...
I agree: considering their resources, they are moving faster than they should. I don't think that Reiza is a big studio, nor do I think AMS2 is a cash cow. It's really a "niche" title. When I look at the amount of content included and the improvements made since the initial 1.0 release, I'm more than happy the development of the sim.
 
Last edited:
J
This is great news. Also, i hope they can fix the raining coming inside some cars like the sprint car. ;) :geek:
 
G
jinxjx said:
This is great news. Also, i hope they can fix the raining coming inside some cars like the sprint car. ;) :geek:
Click to expand...

Yes and when you drive over a hill in rain, that you still see the splashing water of the rain through the hill.... Would be nice to see that fixed after so long, it really ruins the experience.
 
T
Kingkoenig said:
I agree: considering their resources, they are moving faster than they should. I don't think that Reiza is a big studio, nor do I think AMS2 is a cash cow. It's really a "niche" title. When I look at the amount of content included and the improvements made since the initial 1.0 release, I'm more than happy the development of the sim.
Click to expand...
They are not a big studio at all and if you gathered a list of improvements they have made and what is offered in terms of features, content, game modes etc it is much larger than what other games have offered in the same time since their release.

Obviously other sims are more mature than AMS2 in terms of age, but none the less if you think about how quickly Reiza have brought custom grids to the game compared to R3E, or UI updates and improvements compared to rF2, the difference in delivery is not even close.

I'm not trying to talk **** about other developers, and the state AMS2 released in was very rough so improvement was inevitable. I'm more just pointing out the irony when I see people complain about the developement speed of AMS2 when in comparison to the rest of the market they are market leaders IMO.
 

Latest News

Article information

Author
Mike Smith
Article read time
2 min read
Views
11,742
Comments
70
Last update
Join the discussion

Share this article

Are you buying car setups?

  • Yes

  • No

Results are only viewable after voting.
See comments…
Back
Top