DiRT Rally’s Sideforce has No Simulation Value

sweden_i20_2_rEver since we established DiRT Rally as the greatest racing simulation of 2015 here on PRC.net, a small yet extremely vocal crop of sim racers have cried foul. Despite our gushing review of the title in the fall of 2015 – which indicated you needed to run out and purchase this game if you hadn’t already – certain hardcore sim racers were simply not sold on the potential this game had to dethrone Richard Burns Rally as the greatest point-to-point off road experience of all time. For a while, I believed these anonymous users were nothing more than obsessive trolls, desperate to extend the lifespan of a simulator that they had grown emotionally attached to, rather than move forward into something new.

Make no mistake, DiRT Rally is a phenomenal rally game. The combination of impressive visuals and authentic physics is something the sub-genre desperately needed, especially considering front-runner Richard Burns Rally had been conceived during the height of the PlayStation 2 era and most certainly wasn’t easy on the eyes. However, when it comes down to comparing the two titles in terms of accuracy in the raw driving model, dedicated DiRT Rally players have accidentally stumbled into the lone element keeping DiRT Rally planted firmly on the simcade side of the fence.

Downforce is the process of large quantities of air being displaced over top of a race car while at speed, which plants the car to the ground and increases the vertical force on the tires – producing an increased level of grip. Sideforce, on the other hand, is the process of airflow catching the side of the car while the vehicle’s angle of attack is not directly facing forward; in simpler terms, air hits the side of the car while cornering and also generates downforce. The reason you’ll see World of Outlaws Sprint Cars run such mammoth asymmetrical wings, and retain ludicrous speeds through the center of the corner, is to generate insane levels of sideforce and drive the whole track almost wide open.

006af168e164a3fbfae4e67764d90823There have been a group of sim racers claiming DiRT Rally’s tire model has been heavily simplified for mass market appeal – it’s a Codemasters game after all – but to everyone’s collective surprise, this isn’t actually the case. The main problem with DiRT Rally’s physics, as displayed in the video uploaded by Ethan Dean above, is the ridiculous levels of sideforce generated by every car in the game. Dean can be seen throwing a variety of rally cars off of the biggest natural jump in the game, and they are literally curving in the air like a cheeky baseball pitch. The Peugeot Group B entry, which was literally just a hatchback with an absurdly powerful engine stuffed in the boot, takes off from the far side of the screen in the very end of the video, and curves so rapidly mid-air it actually goes into the spectator section with what appears to be forward propulsion.

side3Technical terms aside, DiRT Rally’s physics engine, and the underlying values governing each car, have been written in a way where there is a literal hand of God catching the car in the center of the corner and stabilizing it far beyond what would ever occur in real life. This is what’s allowing some of the Aliens on YouTube to post ludicrous world record times while seemingly throwing the car around like it’s an Xbox game from a decade ago. Codemasters have given econoboxes with no aerodynamic pieces whatsoever, the aerodynamic properties of a Sprint Car when cornering.

The delicate act of balancing the car on the edge that has become a stable of Richard Burns Rally’s grueling difficulty level, is simply not present in DiRT Rally. You have a giant invisible wing guiding the car when you’re even the least bit sideways.


70 thoughts on “DiRT Rally’s Sideforce has No Simulation Value

  1. lol and people still believe this guy anytime he screams this game has simulation value this game doesn’t have. This game has countless bugs, this game doesn’t.

    He gets all content from other people. For example he never writes his own opinion or findings about a new update, he always goes to the forum to see what people say negative and consider it truth. Then anything praising it is discarded or demeaned.

    Oh you need to change this value on AC’s tyres, then we get simulation value, oh you need to not copy paste values, then we get simulation value, oh whatever the physics guru say and we get simulation value. Oh the bmw has a big aero geometric figure in front and in the back of the car, no simulation value.

    Stop doing disinformation dude.

    Face it, you can’t identify whether something is right or wrong in terms of simulation physics. So please stop “informing” your readers when you don’t have anything objective to back up what is right or wrong. Maybe you could learn some things from «DrRacing’s blog».


      1. Ive never seen this in the game unless I took a bump before the jump that shifted the vehicles weight causing a spin…that or he turned the wheel which is nearly impossible to see in this video especially if he barely turned it. We also don’t know what tune is on the car, hell he could be doing this with a combo of gas and brakes.


        1. “Ive never seen this in the game unless I took a bump before the jump that shifted the vehicles weight causing a spin…that or he turned the wheel which is nearly impossible to see in this video especially if he barely turned it. We also don’t know what tune is on the car, hell he could be doing this with a combo of gas and brakes.”

          Non of what you wrote has anything what so ever to do with whats being presented,to suggest you can essentially “curve ball” a car over a a small jump with the way you turn the wheel,or “gas brake” combo (?), is like much of what you say, utter bollocks, not trying to be mean,really.


          1. Why are you even quoting if you’re directly replying below him and you’re gonna put his entire text inside the quote?
            We don’t need to read two times the same time, so just quote a specific part or just reply below him.


  2. Is it just some of the cars or all of them? I’ll do some testing when I get home, but I hadn’t noticed this on the Pike’s Peak cars (which do have massive downforce).

    Still a blast to drive but would be great if they could fix this.


  3. I’m inclined to agree with the guy above — I don’t think Austin knows for himself whether this physics error exists or not, he just whipped up an article because someone else posted something about and surely saying something like this about the very popular and well received Dirt Rally will bring “controversy” and moronic comments and page views.

    Also I have to take issue with your categorization of the Peugeot 205 as “literally a hatchback with a big engine in the back.” If you really think that’s the case, you don’t know much (if anything) about the car.


  4. One of the many forums James/Austin lurks (with different name ofc) is this PC Racing Sim Thread on NeoGaf Forums

    Google —> PC Racing Sims Thread – NeoGAF


  5. I think if you take into account all aspects of the game together, I’d easily rank DiRT Rally the best rally game of all time. It feels so weird being positive about Codemasters. Imagine if they took the same approach for F1? People can use the “approachable” card all they want, but proper physics paired with assists have always worked, so that’s no excuse. If F1 could stop one year due to a strike, CM would have two years to switch engines and build from the bottom up.

    I’m not about to reinstall RBR, but I always enjoy being educated. I guess this is the perfect example that things go forward, but sometimes they go a bit sideways too. One game has better stages, another has better feel over certain types of terrain, another one looks better, WRC is shit, etc. Gotta say that video was pretty funny. Perhaps the same hand of God that guides your car also guided the skunkworks at CM so that something good finally came out of that shite factory.

    On a related note, I bought the pCars Renault pack on sale and I gotta say that after tweaking the FFB, the Formula Renault feels much better than the broken one in rFactor 2. Come at me, cavity scratchers.

    Liked by 1 person

  6. Old ass news. We’ve been talking about the weird physics on the forums since the game came out in Early Access. Here’s another shocker, the tarmac FFB and physics are all kinds of fucked up. Always have been.

    On the other hand, I’m sure as hell not convinced RBR simulates anything more, or more accurately.

    Liked by 2 people

    1. Yeah, RBR’s tarmac is about as bad as Dirt Rally’s. At least RBR’s physics is not overly jumpy on the z-axis and isn’t artificially stable on the yaw axis, also the car in RBR is not just a single point in space.


  7. DiRT Rally has always been great fun to drive.. ..as long as you drive front end first as you should drive the real car. When you start pushing real hard the physics engine will show its dirty face and you most likely want to rage quit the freaking game.

    Have fun or compete. The choice is yours.

    Liked by 1 person

    1. To clarify – sure, you should always jump straight, but I only hucked it sideways off jumps to show that the issue of excessive grip is aerodynamic and not down to weight or tire model or whatever. This force acts just as strongly on the ground, and allows you to take corners far faster than you should be able to.

      DiRT Rally would feel pretty much perfect if it wasn’t for this issue, I’m dead sure.

      Liked by 1 person

      1. Man, this sucks. Seems like it would be an easy fix. Any luck getting CM to listen?

        I wonder if they implemented this purposefully to make things a bit easier. Now that you mention it, I do sort of recall feeling something pushing the car back towards the track on long sweepers. I recall having to actually watch it since it can cause you to overcorrect. After a while, though, my brain sort of internalized it and I forgot about it.

        What a shame. I thought I could drive pretty good there and turns out to just be dodgy aero 🙂


  8. Oh well,iracing doesn’t model side force at all so which has more simulation value I ask,given the fact that iracings tyre model is an embarrassment how can it have more,if I were do I’d be hanging my head in shame,makes a tyre model that heats the whole fucking tyre in a lockup,the very definition of pathetic fail.


      1. And who belt up…love those vids on YouTube. Showed my wife one and she just chuckled softly, then walked away.

        Probably the reaction those guys get from women in general, I imagine.


  9. “Downforce is the process of large quantities of air being displaced over top of a race car while at speed, which plants the car to the ground and increases the vertical force on the tires – producing an increased level of grip. Sideforce, on the other hand, is the process of airflow catching the side of the car while the vehicle’s angle of attack is not directly facing forward; in simpler terms, air hits the side of the car while cornering and also generates downforce. The reason you’ll see World of Outlaws Sprint Cars run such mammoth asymmetrical wings, and retain ludicrous speeds through the center of the corner, is to generate insane levels of sideforce and drive the whole track almost wide open.”

    LOL wow that was one funny bit. Does your educated crowd read that and go “hmmm hmmm yes it does indeed make a lot of sense, this James writer dude is certainly one smart fellow.”?


    1. Yes, that’s a pretty poor explanation of what’s known as the Weathervane Effect. It certainly doesn’t produce “downforce”.

      In actuality, cars with a lot of aero grip from airfoils (as opposed to the venturi effect from diffusers and ground effect tunnels, which also don’t play nice with yaw) tend to perform terribly with large yaw angles. You get delamination of airflow and sudden loss of lift. Yes, I said lift. The airfoils are simply upside-down, but it’s still lift.

      None of which has anything to do with weathervaning or this “sideforce” stuff. This looks like a bug.


      1. We live in a world where people use the term “mechanical downforce”, using it as a shorthand for aero grip is relatively harmless by comparison.


  10. A sim that James praised for more than a year suddenly becomes an unsim because of some shit he read yesterday. This is a completely new phenomenon that has never happened before.


  11. I’ve always felt something about the cars’ inertia felt weird in DR and I bet there’s a problem with how they’re modelling that rather than something with sideforce which implies a higher level of complexity in the engine than we should probably be giving them credit for.


  12. To every one with sand in their fucking vagina, over a blog reporting a bug he found interesting (and relevant to sim racing), YOU ARE THE PROBLEM IN SIM RACING, this isnt a attack piece or any such nonsense, its simply highlighting a interesting bug in a decent game, who the actual fuck thought DR was even “simulating” this stuff for real in the first place FFS, as far as I know the only ppl to get close to simming full aero and all its nuances would be ISI, stop upsetting yourselves and getting all fucking defensive, its not changed ANYTHING for those that enjoy it still, stop acting like SMS and Kunos fuckboys, maybe jump on google and expand your knowledge on a very interesting and relevant subject.

    Liked by 1 person

      1. “calm down”

        Im Calm as, fuck you

        “close the pc”

        For a start who “closes” a PC? second this aint 1999, there is far more then one option to “connect”.

        “go out”

        You sound like a spaz, if gonna try be a smart arse to a smart arse I suggest you up your game amigo, like this….

        “Take a deep breath, pull the stick outta ya arse, shut the PC down and go Outside”

        You see, you can get your message across and not sound like a Autistic 12 year old, Id recommend running your lame arse response’s through me first,not only will it help with your lack of education but more importantly your wit, or lack off.


        1. Hash stop being a little gay cock loving bitch and throwing your toys out of the cot, calm down, its just the internet you prolapsed little knob jockey


        2. You have serious problems Hash, you sound like an autistic muslim who just found out the 72 virgin story isn’t real, whats wrong with you?
          Is your anal cavity hurting again?


    1. lol @ thinking DR is even remotely close to simulating any of this stuff. All DR is doing is : Hey the car is pointing in THAT direction, let’s move it THAT way. Cause physics.

      If DR was giving a car more downforce with increasing yaw angle, the car sideways in the air wouldn’t fly in an arc towards the sides, it would fall DOWN quicker. lol “sideforce”.

      All this “discovery” shows it that DR physics don’t make any sense. A thing anyone with a little bit of common sense has figured out after 10 minutes of driving on them.


  13. > “Ever since we established DiRT Rally as the greatest racing simulation of 2015 here on PRC.net”
    I choked on that bit. Well, that kind of explains why you keep R3E in such a high regard, James… It has nice sounds 🙂 If Ian Bell wouldn’t act as a massive dick, I guess Project CARS would be considered yet another greatest racing simulation here at PRC… Also, if FH didn’t have all the issues.

    Why don’t you simply ignore the whole “simulation value” thing altogether and just write about technical issues in modern games and how they are being sold with no QA work done on them, James? Physics is just not your thing, and you have no interest in FFB either. You are not interested in simulation, you are interested in gaming, so quit pretending you care about realism a lot.

    Also, that would enable you to speak more openly about any genre, not just simracing. You tend to spam every second article with Madden or any other off-topic stuff that comes into your mind during writing the article anyway, so make it a blog about the modern video games industry instead. This might also make the place more popular among other folks too.

    As for correct lap times in a title, you can always tune up Mario Karts cars to reflect real life ones. Won’t make their physics any closer to reality.


  14. Also, what happened to “The truth is that Milestone built an absolutely phenomenal rally simulator, one which blows away both Richard Burns Rally, as well as the more recent DiRT Rally in terms of both driving physics and content, but the entire experience is plagued by absurd technical problems that make the game tread dangerously close to unplayable territory.”? The article says “James” next to it. Or did you write SLRE off simply for having seemingly more issues than Dirt Rally has? And for the sound, of course, too.


      1. So are Dirt Rally’s. One thing is these two titles’ bugs, their physics and handling is completely another though. Yes, DR presents a more sound overall package, but that certainly is not enough to secure the title of “the best racing simulation of year such and such”. It has to be a simulation first and foremost. And then it comes to being a simulation, Project CARS looks actually more of a sim than Dirt Rally. I’m not even talking about tarmac here, just in general.

        Speaking of Project CARS. I suppose you remember WMD gathering funds for the development of the second installment. The lowest tier was around $60, if I remember correctly… Do you have any info on what happened to that?


        1. Project Cars is on the same level as Gran Turismo and Forza when it comes to simulation. It is a mediocre broken simcade racer that is supported by a weird cult of SMS backers. Dirt Rally beats Project Cars on physics alone, and we haven’t even gotten to the other features yet.


          1. So is Dirt Rally. Project CARS is broken, but at least it doesn’t feel like the car is a point in space there. Also, it doesn’t seem to be having some sort of yaw rate dampening going on as Dirt Rally does. And on tarmac it’s absolutely ridiculous. Not even simcade level.


          2. Another thing. In case you thought I was an SMS fanboy, I have both titles uninstalled. Neither one is worth wasting my time. But if I had to name one of them a sim, it’d be Project CARS.


  15. I’d be genuinely interested, not just for this particular phenomenon, in the full lists of a) wheel settings / advanced wheel settings, setup, trace of the steering input, driver settings (esp. DOR is impottant, because the smaller you make it to the car’s default, the easier it basically gets, because you can give much greater input per movement (throwing around, say, the S4, with wheel on 900 degs (or was it 720,, I don’t know from memory) is _hard work_ ) (*). Was manual shifting with clutch done (another thing that makes a rather massive difference, simply in amount of body movements you need to time and coordinate? Then, what was the used setup? What’s the input from wheel, pedals, gear, clutch, what’s the line driven, were any obsracles used as lau chpad, etc.

    If we have that, we can all try and reproduce the behaviour on a) their settings, b) your personal settings, c) factory default settings. Then we collect the data, have a good hard look at it, and will at least have some basis in shared experience to talk about, instead of talking in completely undefined terms like ‘the physics’ as if that has any generally accepted, testable and verifiable meaning – I think it’s a weasel word, actually, butthar’s just my opinion, man.

    Because that’s how you do proper empirical research. You gather dara, carefully controlling certain variables while keeping others the same. And then you have different people who all agree on the concepts and protocol used to do it. And the you have data, patterns, occurrences, etc., a wealth of information onwhich to base your criticism.

    Because without, it can be anything … simulations are works of compromise with many complex coupled differerential equations (pi) that need to be numerically solved in real time and which are by necessity only valid within certain boundaries, in order to keep them solvable to the desired accuracy. I don’t know who of you hassome experience with complex systems and numerical similation, but ir’s quite easy to hit a particularly instable volume in a particular model’s parameter space where very unpredictable things start happening, even more so if you tinker around with mods, or otherwise ‘configure your sweet spot’ of wheel settings, FFB, and car setup to be able to make inputs that although possible within the model, result in wildly unpredictable inputs to the model.

    Starting to see what I mean? Try it yourself. Even PC became astonishingly fun ootb when I stopped tinkering and reset everything to default.
    ‘Detail’, ‘feel’, ‘nuance’, …. all there, strangely.

    Anyway, just my early rainy Sunday morning in Berlin thoughts. If any one of you is genuinely interested in seeing if we can at least informally do some research on reported bugs/issues, the way I described, in a genuine attemptto get a more tangible description, feel free to contact me. I can write the protocol and do the data analysis – all you have to do is drive and be dispassi8nate 🙂

    Cheers, Constantijn

    (*) Automobilista overrides my driver default DOR to 450. I don’t know if it’s a known bug or if I just miss something obvious. It doesn’t bother me too much, since I now know it’s there, so I always check. Anyway, I first noticed it when I booted up DR and started a Daily in the 2010 Mini. I’m familiar with both the car and the stage (Wales, iirc) but to my surprise steering felt tighter and I was flying – beat my own stage best by a couple of seconds which for me is too big a discrrpancy for it to just have been a ‘good day’. It really felt weird though, and when I checked everything, it turned out I was, unbeknownst to me, at the aforementioned 450 DOR. It makes sensethat it’s easier the smaller the DOR, and everything else staying the same, the ffb has way less range to transmit its torque/time, hence my ‘squashed’ metaphor.


Ratio of vowels to consonants will be monitored. Post at your own discretion.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s