Sim Discussion Monday - Assetto Corsa

[Paul Jeffrey]

In today’s edition of ‘Sim Discussion Monday’ we talk about all things Assetto Corsa!

Love it or loathe it Assetto Corsa has quickly become one of the most popular racing simulations of the modern sim racing landscape. Bathed in exotic licences the likes of Ferrari, Porsche and Lamborghini all pitched together on laserscanned tracks, with high quality graphics and physics and a vibrant modding scene, Assetto Corsa from Kunos Simulazioni has very quickly picked up the mantle once held so tightly by Image Space Incorporated and the venerable rFactor platform.

However despite the popularity of the software Assetto Corsa still lacks many of the features considered essential in a modern racing title (think day / night transition, weather and driver swops), leaving many to feel the sim is so close to the top of the mountain, yet just falling short as it scales the summit.

Now in its third year of development, let’s use this article as an opportunity to discuss the game, share tips on optimising the experience on screen and through the wheel, highlight some great content and basically just talk about anything and everything Assetto Corsa!

Mondays be like...

 

[StefanoCasillo]

Thank you, sir. Tells me I'll have to adjust my FFB so I can feel it.

also we model steer assist on modern F1 and LMP which is designed to "compress" the signal and reduce the differences between high and low load... cars without it (ie. the Lotus 98) should show a more pronounced difference.

 

[mms]

this is the FF signal at 62 kmh turn on the F138, it reads 49% :

View attachment 207032

and this is the reading at 202Kmh.. 99% so twice as that.

View attachment 207033

I was going to make a video with the exact same combo using the pedals app (unfortunately FFB does not get saved in replay, otherwise it would be already on YT)

 

[Jim Liak]

[QUOTE="neuer31, post: 2552430, member: 113561] Βut from a pure physics standpoint Assetto Corsa is inferior.
[/QUOTE]

Says the guy who plays a game that will not support dirt on tyres...
As mentioned, only the guy that will have on his hands all the code of all sims will know which is the best.
Then again all sims trying to simulate.
Devs have chosen to simulate different aspects of life better and other more basic.

A really nice experiment would be a modder that has access to all the data of a car, to try and make the car in all platforms and then compare which is closer to real one.
@mclarenf1papa comes in my mind with his Perrinn car.

 

[Will Mazeo]

meh if people want to have an idea about what sims simulate just open a physic file of the cars in the sims that allow you to do it. If you want to go ahead contact some GT3 teams and get data, many of them are very friendly...
The ones that don't let you see the physic files... oh well, we'll never know if they are adding F1 numbers to make a Porsche Cup ¯\_(ツ)_/¯

 

[St3fan]

It should be made an industry standard that racing simulator loading screen must show a rating indicating its simulation value and superiority, so we can avoid fruitless discussions about whether the one I'm playing is better than yours.

Ratings shall be determined by an Internet sim racing physics experts board.

 

[snyperal]

I watched empty boxes comparison of the porsche cup in iRacing, Assetto corsa and the fake one in automobalista, interesting watch. I think the only difference was the fuel consumption was higher in AC. I agreed with his sentiments that 3 developers can use totally different methods to create one car insert that into their simulator and produce a very believable, similarly handling vehicle with very close to each other and RL porsche laptimes. What a world we live in! A few years ago you would do well to recognise the car or model in a game or sim, now we're splitting laptimes and fuel consumption as comparisons.

I don't care where they get their numbers from it the end result is that!

 

[RMR1]

I watched empty boxes comparison of the porsche cup in iRacing, Assetto corsa and the fake one in automobalista, interesting watch. I think the only difference was the fuel consumption was higher in AC. I agreed with his sentiments that 3 developers can use totally different methods to create one car insert that into their simulator and produce a very believable, similarly handling vehicle with very close to each other and RL porsche laptimes. What a world we live in! A few years ago you would do well to recognise the car or model in a game or sim, now we're splitting laptimes and fuel consumption as comparisons.

I don't care where they get their numbers from it the end result is that!

You know, before AC, this view of things got you burned on a pile!

 

[Amir Vodokotlic]

Comparing the 650s GT3 and 650s GT3 in AC and rF2 I can't feel the difference.

They drive the same to me. I am assuming the iRacing 12C GT3 is probably same to AC/rf2.

I really don't know where some of you get the one sim is better than other sim argument. Without having access to real data to compare.

Unless by some miracle all of you're data engineers that pull telemetry and analyze it deeply to say one is right/wrong.


IMO Only thing that can ruin the experience between the three games is ffb for me.

If you can't set it up right (FFB) it can ruin the feel and than you go on forums and say x > y or vice versa.

 

[tlsmikey]

I have a question about AC going forward. What are the next enhancements likely to be? It seems like there was a pretty rapid development pace until 2017 and now things have slowed dramatically. To the point, where i'm wondering if they're working on a sequel or something else.

Are we basically waiting for a ferrari pack and that's about it? The car packs no longer do it for me and I find myself at a place with AC where I'm not sure what else i'm missing that can be addressed with the current framework.

 

[David Dominguez]

rFactor 2 has (from the released Sims) the best physics by far atm. Maybe Project CARS 2 will be better (cause Livetrack 3.0). Both are calculating the physics dynamically. Assetto Corsa uses table based physics which is massively inferior (it can produce good results though, Assetto Corsa is a very good example of that).

Ironically, rF2 tire model uses insanely long lookup tables, pre-cooked with a (virtual) tyre tester software.

 

[Patrik Marek]

I won't say that your opinion is wrong, but I'll give you an example that happened to me last Sunday. Around 14:10 the race starts and I'm the blue Alfa that is in 5th, when the green light appeared, someone slightly touched my car, and you can see how shaky I was when that happened.

I "like" how so many people say how Safety/Pace car is waste of time and not needed, and then in leagues you see that people actually drive one of the cars for this purpose ..

 

[Patrik Marek]

If you can't set it up right (FFB) it can ruin the feel and than you go on forums and say x > y or vice versa.

and this is what always makes me laugh
different FFB settings doesn't change the physics, doesn't change how the car moves on road and reacts to inputs, .. it only changes how it FEELs and that's subjective thing

which is why I always find these debates amusing, people are comparing what is better simulation, but rather then the actual numbers, measuring how much cars moves around etc. , they just go by the feel of the FFB, which to me is absolutely wrong approach

there was a video on YT, where guy was studying a car behaviour in iRacing, based on measuring the distances it travelled , speeeds etc, and form there he was making his point that certain behaviour was wrong in iRacing, becasue car was producing unrealistc results

that to me is looking at sim properly,

not based on hwo the plastic ring in my hand vibrates or not, evenmore so when you never sit in actual race car and have no idea what you would feel through steering wheel or not ...

 

[Deleted member 113561]

@Patrik Marek: I agree with you. Its not only at such sim discussions here, its a general problem. People use facts not enough in discussion (generally). In extreme you can see that at conspiracy theories (ie. 911) - from believing total misinformation (unintended ones) up to believing straight up lies - cause they don't check source or do their own research.

@David Dominguez: A rF2 dev said that once that rF2 is not table readout (but rF1 is) - mmhhh I hope I can provide you source to it, gonna look for it.

mod-edit: cleaned up

 

[Tomi]

Of course avoiding answering such questions so your fans won't be disappointed that they don't have the best physics. I don't say its not a good game though.

How about you give us all a deep analysis and explanation why those two games you mentioned have the best physics. And please use some arguments other than their marketing Info what you can find just by going to their web sites.

And I'm not trying to defend AC or claim they have the best physics because I really don't know which one have. Just enjoying AC the most because it feels best for me.

 

[David Dominguez]

@David Dominguez: A rF2 dev said that once that rF2 is not table readout (but rF1 is) - mmhhh I hope I can provide you source to it, gonna look for it.

Straight from ISI document:

Lookup tables aren't a bad thing; very likely even F1 teams use them.

 

[Emery]

@David Dominguez: A rF2 dev said that once that rF2 is not table readout (but rF1 is) - mmhhh I hope I can provide you source to it, gonna look for it.

@neuer31 - rF2 is my #1 sim and I'll back it when I can. However you obviously don't know what you're talking about. David is absolutely right in that the TTOOL which a modder uses to build tires creates a lookup table. The lookup table describes the contact patch and sidewall flex, the overall shape, from different loadings & camber, which is tied to the graphics. rF1 uses a lookup table for the shape of the slip curves (there are more than one) and intermediate values for the slip curves are interpolated from those tables. The slip curves are still used by the AI in rF2.

TTOOL easily takes overnight to build up that table and, if the high precision option is turned on, even longer. This is why tables are used--the calculations cannot be handled in real-time. While much of rF2's heritage descends from rF1, the TTOOL technology is taken from rF Pro. Animating the tire's behavior to match TTOOL's output has still not been equaled, though pCARS 2 has attempted (the one animation from pCARS 2 that is on YouTube incorrectly does it, by animating the entire tire with the contact patch behavior... I don't know if SMS will put tire flex animation into the game).

 

[Tomi]

rFactor 2 has (from the released Sims) the best physics by far atm. Maybe Project CARS 2 will be better (cause Livetrack 3.0). Both are calculating the physics dynamically. Assetto Corsa uses table based physics which is massively inferior (it can produce good results though, Assetto Corsa is a very good example of that).

rFactor Pro - rF2 is a cutdown version of that. But in its core this sentence is absolutely correct.
But don't tell that iRacing or Assetto Corsa fanboys ...

The technology in rFactor and PCARS (especially 2) is massively superior to every other sim out there at the moment (why was already mentioned). It doesn't mean its a better experience though, that you always have to keep in mind! But from a pure physics standpoint Assetto Corsa is inferior.

And after those messages you write that one below

I agree with you. Its not only at such sim discussions here, its a general problem. People use facts not enough in discussion (generally). In extreme you can see that at conspiracy theories (ie. 911) - from believing total misinformation (unintended ones) up to believing straight up lies - cause they don't check source or do their own research.

 

[Durge Driven]

So many people mentioning the lack of dynamic weather system and rain...

In rF2 I bet most of you guys never or very rarely race in the rain (just like myself)..

+1

I don't see rFactor2 rain as very dynamic or a strong point over Assetto Corsa and I'm a fanboy

Working wipers and no rain on windscreen, wet tyres missing, no running water, puddles , spray is basic, no dynamic wind so rain always falls straight down even in a storm
Drying line is linked to real road rate which can cause track to never dry or dry too fast
bugs where rain can get grippy

I think like most that is insignificant to me compared to rF2 dry weather running

 

[Patrik Marek]

+1

I don't see rFactor2 rain as very dynamic or a strong point over Assetto Corsa and I'm a fanboy

Working wipers and no rain on windscreen, wet tyres missing, no running water, puddles , spray is basic, no dynamic wind so rain always falls straight down even in a storm
Drying line is linked to real road rate which can cause track to never dry or dry too fast
bugs where rain can get grippy

I think like most that is insignificant to me compared to rF2 dry weather running

that's where you need have it done properly - pCARS2, - you will love that!
and then still race in dry most of the time

 

[HypoToad]

Lookup tables aren't a bad thing; very likely even F1 teams use them.

RasmusP's FFB LUT has transformed my Driving Force GT wheel!