How Popular Are Our Favourite PC Racing Sims?

Most Popular Racing Sims 01.jpg
Racing simulators tend to have more longevity than most sports games, but how many people are still playing some of the bigger titles?

Using SteamDB, we can get a reasonably accurate gauge of how many people are currently playing a particular game through Steam. Below I’ve noted the all-time peak concurrent players for some of the more popular titles here at RaceDepartment, plus the 24 hour peak as taken on November 23rd, 2021.

NOTES: SteamDB of course excludes console players and those who might be using an alternative launch method such as Epic. There is also some level of error both in the statistics gathering, and the times when the data was collected. So, please don’t treat the figures below as an absolute value of how many people are playing the title, but rather an overall barometer of how many people are enjoying it on Steam. iRacing statistics are harder to come by, but I will add a footnote if I can gather an accurate player count.

Assetto Corsa
  • All Time Peak – 11,691
  • 24 Hour Peak – 6,824
This modding monster continues to enjoy success eight years after it was first released to the public, thanks in no small part to the great community of creators that keeps AC relevant with new and classic cars and tracks, plus the latest skins of our favourite drivers and teams.

Assetto Corsa Competizione
  • All Time Peak – 6,926
  • 24 Hour Peak – 3,959
The official SRO simulator has been another success for Kunos. On the eve of the release of another major update, thousands of players are still enjoying this GT sim two and a half years after its release on Steam.

Automobilista 2
  • All Time Peak – 1,150
  • 24 Hour Peak – 372
Reiza Studios are hard at work on their biggest physics update so far for AMS2, which is set to be released to the public this month. The title has had its share of doubters, but Reiza has kept its ear to the ground in the sim community and has made major strides since the first release version and will hope to see the number of people enjoying the title continue to climb.

BeamNG.drive
  • All Time Peak – 13,572
  • 24 Hour Peak – 7,932
Known mainly to the layperson as a crash simulator, BeamNG is perhaps better summarized as a physics playground with a huge modding community. This simulator has more players than even Assetto Corsa, and likely owes much of that popularity to the huge selection of car and track mods on offer, much like AC.

F1 2021
  • All Time Peak – 24,513
  • 24 Hour Peak – 11,400
The official game of F1 has enjoyed mostly positive reviews once again this year. And unlike recent years, updates to the title have included new content. Three additional tracks have been added post-release, which seems to be having the effect of sustaining the interest of players, as over 10,000 people on Steam alone are still enjoying playing F1 at any time.

Euro Truck Simulator 2
  • All Time Peak – 61,081
  • 24 Hour Peak – 37,635
This is obviously not a racing sim, but for perspective on the other titles, check out the impressive numbers behind ETS2. Apparently the allure of a chill drive across some of the most beautiful areas on the continent of Europe holds great appeal to gamers.

Most Popular Racing Sims 03.jpg

Forza Horizon 5
  • All Time Peak – 81,096
  • 24 Hour Peak – 37,924
Forza Horizon 5 has broken every sales record for a racing game, and even ignoring the massive player base on console, the concurrent player count is staggering. FH5 continues the series’ open world, consequence free feel, and appeals to the fun-chasing side in racing game fans.

NASCAR 21: Ignition
  • All Time Peak – 323
  • 24 Hour Peak – 42
NASCAR 21’s launch was marred by bugs and bad reviews. As such, the game is off to a slow start. This is the newest title on this list, and should be enjoying its “honeymoon” sales phase, but most of the major complaints about the title have yet to be addressed. The good news? It’s now playable for me since the latest update, so my overdue review should be coming soon.

RaceRoom Racing Experience
  • All Time Peak – 2,630
  • 24 Hour Peak – 1,057
There’s a case to be made that RaceRoom is the most complete racing sim on the market. As such, its popularity has remained strong since its early 2013 release. Content releases have slowed recently, but a well-built multiplayer system keeps players coming back.

rFactor 2
  • All Time Peak – 1,652
  • 24 Hour Peak – 747
rFactor 2 has been praised since its 2015 Steam release as the pinnacle of simulator force feedback and vehicle physics. Either opinion is debatable of course, but there’s no doubt that the driving experience in rF2 can be a pure pleasure. The title is also the preferred sim of many major eSports series, which keeps it relevant and popular with both casual and professional players.

Obviously there are countless other titles that could have been included in this article, so if I missed any titles that could provide interesting data, be sure to share the name and user stats below.
  • Like
  • Haha
  • Love
Reactions: James Cook, Knoxstar, Steve007 and 25 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 Other Games

Comments

M
Neils on Wheels said:
Honestly a lot of waffle saying absolutely nothing at all.

Sim (simulated) racing or racing simulation are the collective terms for racing game software that attempts to accurately simulate auto racing, complete with real-world variables such as fuel usage, damage, tire wear and grip, and suspension settings.

Forza horizon is a casual arcade racing game design for a game pad aimed at the masses & doesn't fit into the category of sim at all.
Click to expand...
Just saying, but Forza Horizon does have damage, tire wear and advanced car setup options. Their physics engine is way better than what is being used in other arcade racing games.
 
RCHeliguy
Premium
Kyuubeey said:
Simulating reality in its entirety isn't the point of a model either. Think about what the word means.
Click to expand...

There are many ways to skin this cat.

In the RC Helicopter Simulation world, there were a few simulators that basically felt right at higher head speeds, but were all just approximations and adjustments coming from a common model.

Then AccuRC came out. They actually simulated airflow around the blades in 50 locations. They simulated the complete flybarless configuration and worked based on the characteristics of different servos and motors etc..

There were hiccups, and initially the CPU/GPU requirements for it were pretty high. But it was better and over time they make the physics engine more efficient. Their physics model kept improving and as a results they could make accurate models of new RC helicopter designs and have them behave much like the real thing.

My only point is that a real simulation of the physics will eventually be the best way to model something. This is complex and it takes time to understand all the factors that are at play in the real world, so you end up with an iterative process as you learn and understand more.
 
H
Kyuubeey said:
That's all fine and good to say if you don't actually know what the delta is from expected outputs. It's largely car specific too.

For years people have been saying stuff like elastokinematics are the end all be all or that simulating every tire flex properly is the end all be all; then I talked to people who did those and it's really not what it's made out to be. So don't make big claims if you're not sure.

Probably the most influential out of all the 'huge chunks' is having a chassis flex model and that only in some cars. Majority of users would not tell any difference when driving a roadcar with one and without one.

Simulating reality in its entirety isn't the point of a model either. Think about what the word means.
Click to expand...
On the point of chassis flex, i don't know about the "driving experience", but I had to work on an AI algorithm for a Lamborghini Huracan GT3 with "LP Racing" team (a really s**tty working experience, but at least I've got my hands on the car!), and the telemetry guy told me that an important parameter for my AI would have been how many km the chassis had, because its flex would vary over time and it was something that had a visible effect on telemetries.
 
Kyuubeey
hotak said:
On the point of chassis flex, i don't know about the "driving experience", but I had to work on an AI algorithm for a Lamborghini Huracan GT3 with "LP Racing" team (a really s**tty working experience, but at least I've got my hands on the car!), and the telemetry guy told me that an important parameter for my AI would have been how many km the chassis had, because its flex would vary over time and it was something that had a visible effect on telemetries.
Click to expand...
Yep, it is not something you can ignore in GT3 cars. Nor in LMP2 cars. That's what I've been told at least; your roll stiffnesses can be very, very different from a 100% rigid single rigidbody kind of setup. That's probably one significant reason for "kerb of death" and why ACC implemented their own method for it.

You can somewhat bake it into the (typically ARB) installation stiffness but it won't be 100% proper. Still a whole lot better than not, though.

RCHeliguy said:
There are many ways to skin this cat.

In the RC Helicopter Simulation world, there were a few simulators that basically felt right at higher head speeds, but were all just approximations and adjustments coming from a common model.

Then AccuRC came out. They actually simulated airflow around the blades in 50 locations. They simulated the complete flybarless configuration and worked based on the characteristics of different servos and motors etc..

There were hiccups, and initially the CPU/GPU requirements for it were pretty high. But it was better and over time they make the physics engine more efficient. Their physics model kept improving and as a results they could make accurate models of new RC helicopter designs and have them behave much like the real thing.

My only point is that a real simulation of the physics will eventually be the best way to model something. This is complex and it takes time to understand all the factors that are at play in the real world, so you end up with an iterative process as you learn and understand more.
Click to expand...

You're not wrong; but don't compare a tire simulation to an aerodynamic simulation. They are not the same thing.

The best CFD on the planet will give you 30% error for a simple scenario; a typical empiric tire model will give you maybe 5% error for a simple scenario. The best CFD on the planet will **** itself when the scenario becomes very difficult; a typical empiric tire model might not even get that bad if the extrapolation is good. Both are wrong but in different ways.

Comparing cats and dogs here, apples and oranges.

Although I've seen 5-7% error from CFD on missiles at best so for that it's quite good from what I can tell. Maybe alright for some aeroplanes too. Godawful for cars, drones, helicopters etc. The more you know the more you realize just how bad it can get but if you cherrypick enough you can fool yourself into thinking it's accurate. Same thing does apply to tire models too...
 
B
Damage Inc said:
I've been saying similar thing, NO racing game simulates reality (as in it's entirity), even the best "sims" leave out huge chunks of reality that can seriously affect how a vehicle operates. So all this dick wanging people do over sims always cracks me up.
Click to expand...
"Dick wanging"....this kind of nonsence always cracks me up.
Here we go again:
Pro engineer's thoughts on a commercial sim:
drracing.wordpress.com

Vehicle dynamics

Again, a very long time i don’t write here. Again, although it was already at a good point, the last project i started didn’t go to an end. I guess that’s life. During the last mo…
drracing.wordpress.com drracing.wordpress.com
And here we go again:
Former F1 engineer using rF2 for professional driver development.
It's not rFPro, it's not just for "visualisation", it's a "game" called rF2. Yeah, I know - horrible UI, long loading times, it's dead.

End of conversation if you know how to read/listen.
My mistake when joining this forum was the impression I'm sharing the same passion with serious adults. It appears it's infested with people who treat this passion as an infantile form of entertainment. Even those with high-end gear are not interested in real motorsport aspects, the driving perfection, tire managment, making good setups.
And all these false claims are starting to make perfect sense now.
I guess if we were to have a gathering in real life, I think I would've find maybe 5 persons around here who are really passionate about racing, and whose childhood dream was to be a real racer.
 
Last edited:
Kyuubeey
I mean it's fine to have whatever opinion and view if you can back it up with at least a hypothesis but it's a bit weird why the less car sim development experience someone has, the more radical their opinions and the less justifications are given.

The only people who say sims are completely unusable/absolutely perfect are people with 0 hours under their belt. :rolleyes:

Then when you have ten or perhaps twenty thousand hours of dev, you're supposed to nod and agree with people who have 1000 or less or you will get the wrath of the whole simracing community directed at you. The more I've tried to steer away from assumptions and opinions and rely more on personal or 2ndhand experience, the stronger the wrath becomes. No wonder there is almost no public discussion about sim physics.
 
Damage Inc
Kyuubeey said:
Simulating reality in its entirety isn't the point of a model either. Think about what the word means.
Click to expand...
But 100% of dick wanging "my sim is better than yours" is based on how Sim (A) is perceived to simulate reality better than Sim (B). They all have strengths and weaknesses depending on the expertise of the developers.
 
Damage Inc
bozakn said:
"Dick wanging"....this kind of nonsence always cracks me up.
Here we go again:
Pro engineer's thoughts on a commercial sim:
drracing.wordpress.com

Vehicle dynamics

Again, a very long time i don’t write here. Again, although it was already at a good point, the last project i started didn’t go to an end. I guess that’s life. During the last mo…
drracing.wordpress.com drracing.wordpress.com
And here we go again:
Former F1 engineer using rF2 for professional driver development.
It's not rFPro, it's not just for "visualisation", it's a "game" called rF2. Yeah, I know - horrible UI, long loading times, it's dead.

End of conversation if you know how to read/listen.
My mistake when joining this forum was the impression I'm sharing the same passion with serious adults. It appears it's infested with people who treat this passion as an infantile form of entertainment. Even those with high-end gear are not interested in real motorsport aspects, the driving perfection, tire managment, making good setups.
And all these false claims are starting to make perfect sense now.
I guess if we were to have a gathering in real life, I think I would've find maybe 5 persons around here who are really passionate about racing, and whose childhood dream was to be a real racer.
Click to expand...
Yes yes yes I own all the top sims, have a rig etc etc......but

I didn't realise that the site was called "Sim Elitism Department".

You need to get these news stories deleted then:


Race Department is for racing games.
 
Kyuubeey
Damage Inc said:
But 100% of dick wanging "my sim is better than yours" is based on how Sim (A) is perceived to simulate reality better than Sim (B). They all have strengths and weaknesses depending on the expertise of the developers.
Click to expand...
Simulating reality *in its entirety*. Think about what that means. Some day we could simulate every atom and probably branch into subatomic stuff too; but not quite yet. Good thing you don't *need* to in order to get a quite reliable model.

It's more about how significant those advantages and disadvantages are when it comes to making an accurate model. People have very faultly perfections about that subject IMO. Mostly people who have no faculties whatsoever to even be making claims about the subject. Opinions change dramatically once some experience is acquired.
 
H
I think that most people not having at least a math/physics/engineering degree cannot really understand how complicated the differential equations describing reality are and how many unknown coefficients they need, and cannot possibly understand what does it mean using partial differential equations to actually solve the problem in real time and how this is already an approximation that may or may not work.

At the same time, not-engineering STEM people may accept a bit less the gross approximation that engineers make every time they have to deal with real-world problems and how approximate methods can get really close to correct results, but often just in some restricted "use-case" where some arbitrary hypothesis holds, otherwise the approximation could even be VERY wrong.

we don't get to know the hypothesis or assumptions the various devs made when creating their physics engine, we may not even be able to see where if fails just observing the telemetry data, but for sure most sims get quite close to reality when you not overdrive the car, and i tend to have a preference on the feeling (FFB and driveability when overdriving) that comes out from more complex ones (RF2/madness for tires, BeamNG for chassis/transmission). Then that doesn't mean that if I had to test setups for a real car, my preference would have been the same.
 
Last edited:
Damage Inc
Kyuubeey said:
Simulating reality *in its entirety*. Think about what that means. Some day we could simulate every atom and probably branch into subatomic stuff too; but not quite yet. Good thing you don't *need* to in order to get a quite reliable model.

It's more about how significant those advantages and disadvantages are when it comes to making an accurate model. People have very faultly perfections about that subject IMO. Mostly people who have no faculties whatsoever to even be making claims about the subject. Opinions change dramatically once some experience is acquired.
Click to expand...

I wasn't really going on about the atomic level. Even basic air/wind stuff isn't done properly yet (as far as I'm aware, stand to be corrected). Name a sim where wind direction and strength actually does anything meaningfull and truly accurate to a car. What about air temp and density, how many "real road/true track" systems have different temps on different corners. Or how about simulating different types of tarmac and the different grip levels of old and new tarmac? There's SO MUCH not done.

Look I get it that you're invested in the high end understanding of the way simulators work, but my point was the way some simmers use it for "dick wanging".
 
Damage Inc
hotak said:
I think that most people not having at least a math/physics/engineering degree cannot really understand how complicated the differential equations describing reality are and how many unknown coefficients they need, and cannot possibly understand what does it mean using partial differential equations to actually solve the problem in real time and how this is already an approximation that may or may not work.

At the same time, not-engineering STEM people may accept a bit less the gross approximation that engineers make every time they have to deal with real-world problems and how approximate methods can get really close to correct results, but often just in some restricted "use-case" where some arbitrary hypothesis holds, otherwise the approximation could even be VERY wrong.

we don't get to know the hypothesis or assumptions the various devs made when creating their physics engine, we may not even be able to see where if fails just observing the telemetry data, but for sure most sims get quite close to reality when you not overdrive the car, and i tend to have a preference on the feeling (FFB and driveability when overdriving) that comes out from more complex ones (RF2/madness for tires, BeamNG for chassis/transmission). Then that doesn't mean that if I had to test setups for a real car, my preference would have been the same.
Click to expand...
Exactly 99% of simmers don't have a clue, only the few with the required degrees/knowledge can remotely understand. It's just that a LOT of simmers who don't have a clue do most of the dick wanging. I have ALL the modern top sims, LOVE them all, massive respect for ALL the devs busting a gut to do the best they can within the resource constraints they have (money, data and processing grunt). I see NO need for the dick wanging at all.
 
H
Damage Inc said:
I wasn't really going on about the atomic level. Even basic air/wind stuff isn't done properly yet (as far as I'm aware, stand to be corrected). Name a sim where wind direction and strength actually does anything meaningfull and truly accurate to a car. What about air temp and density, how many "real road/true track" systems have different temps on different corners. Or how about simulating different types of tarmac and the different grip levels of old and new tarmac? There's SO MUCH not done.

Look I get it that you're invested in the high end understanding of the way simulators work, but my point was the way some simmers use it for "dick wanging".
Click to expand...
The madness engine does all of that, excluding worn tarmac simulation, wind simulation *should* be easy to do assuming a good aero model, the various realroad simulations should also be easy, but extremely computationally expensive when increasing resolution (what the madness engine manges to do, exspecially with water is astonishing IMHO, also BeamNG has realtime fluidodinamics for rivers and puddles, managing to simulate also boats and airplanes as a byproduct)
 
Kyuubeey
Well the main point here is that having all that stuff isn't necessary to have an accurate vehicle model for driver training or setup purposes. Although I know AC can do multiple surfaces just fine; and it's also kind of pointless to try for 100% accuracy on that because no track in existence has a constant grip level on every surface, at every hour of every day. All tires do not respond the same way to the exact same surface either. This kind of stuff is almost beyond the goals of these kind of models.

This is why I keep saying that please don't make assumptions.
 
N
Damage Inc said:
Exactly 99% of simmers don't have a clue, only the few with the required degrees/knowledge can remotely understand. It's just that a LOT of simmers who don't have a clue do most of the dick wanging. I have ALL the modern top sims, LOVE them all, massive respect for ALL the devs busting a gut to do the best they can within the resource constraints they have (money, data and processing grunt). I see NO need for the dick wanging at all.
Click to expand...
What a horrible turn of phrase.
 
Damage Inc
hotak said:
The madness engine does all of that, excluding worn tarmac simulation, wind simulation *should* be easy to do assuming a good aero model, the various realroad simulations should also be easy, but extremely computationally expensive when increasing resolution (what the madness engine manges to do, exspecially with water is astonishing IMHO, also BeamNG has realtime fluidodinamics for rivers and puddles, managing to simulate also boats and airplanes as a byproduct)
Click to expand...
That what I meant when I said devs have to work within constraints of money, time and computational power. Although the main issue with water and the madness engine was they couldn't do it with the ai (computational limitations) and thus that caused a lot of grief. I'm not 100% sure but I feel that a lot of dev "fudge" things, as long as it "feels" right.
 
Damage Inc
Neils on Wheels said:
What a horrible turn of phrase.
Click to expand...
I only use it because it's a LOT easier to type than "my sim is better than your sim". Everyone knows what it means, and it sums up the type of person who says that succinctly IMO. It causes a lot of arguments, I mean people are even arguing with me and I like ALL the sims......something wrong there.
 

Latest News

Article information

Author
Mike Smith
Article read time
4 min read
Views
24,035
Comments
165
Last update

Share this article

What's needed for simracing in 2024?

  • More games, period

  • Better graphics/visuals

  • Advanced physics and handling

  • More cars and tracks

  • AI improvements

  • AI engineering

  • Cross-platform play

  • New game Modes

  • Other, post your idea

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