I think it was around this time last week, I entered a practice session for an event later in the afternoon on Race2Play, and was promptly congratulated by a fellow sim racer for my performance at Edmonton International Raceway the previous evening. If you’re one of my officially licensed stalkers, or you just can’t get enough of this God forsaken website that you need to follow my various social media accounts as well, you’ve probably figured out by now that I’ve thrown my ass in a real race car this year to make a point: It’s cheaper – and significantly more fun – to go racing at your local short track for an entire 22-race season, than to throw money at optional simulator hardware in the endless pursuit of immersion.
How did this all come to be?
The story isn’t nearly as exciting or significant in the grand scheme of things compared to the articles and videos you’ve seen on prominent sim racers such as Ray Alfalla or Greger Huttu. Even though Edmonton is primarily a city that lives, breathes, eats, and sleeps hockey, we’ve actually got our own little sim racing center on the north end of the city. Run by Amjad Othman, Impulse Sim Racing features ten custom cockpits powered by the commercial rFactor 2 software, along with complete leaderboard functionality, and the ability to organize custom sessions with your friends. Considering I have a pretty ghetto sim setup compared to most of my fellow virtual drivers, I began making weekly trips to Impulse as a way to mess around with rFactor 2 in a rig completely different than my own – triple monitors, Fanatec pedals, Buttkicker audio… The stuff considered high-end on most sim forums, Impulse had ten identical rigs with these components. The owner was pretty cool, the atmosphere of the place was comfy, and it was a fun way to meet other like-minded sim racing nerds such as myself. Away from the absurd forum autism plaguing most online communities, there are some pretty cool people who enjoy these games as much as we do.
Predictably, I made a name for myself in dramatic fashion as being that asshole on the leaderboard with inhuman times. Rather than compete in the organized race nights and blow out the competition to feed my ego & reel in the prizes, I willingly sat out and played the role of race steward instead – which was a lot of fun. Every single event went off without a hitch, and there were one or two occasions where everyone hung around the place until almost midnight – re-watching the entire race from front to back. So before y’all jump on me when I question certain officiating decisions of other leagues, keep in mind that I’ve basically run my own league in a place where on-track incidents could result in very real parking lot altercations, and not just shit-slinging over voice chat.
At some point, a couple of local short track racers from Edmonton International Raceway – including last year’s track champion Justin Horton – walked into the place while I was in the process of running solo laps, and knowing I was a huge stock car fan, Amjad shut down my session and introduced us to each other. After couple of intense 40-lap sprint races at Charlotte… Sorry… Mountain Peak Speedway… I was asked if I’d consider coming out to the track during the 2016 season and driving one of their rental cars for a race or two – a process made significantly easier considering the track owners already knew me, and my NASCAR license was just sort of sitting in my wallet. Realizing someone who actually won a championship and knew what he was doing would be maintaining my car, I upped the deal a bit and asked what running the entire season would cost.
It was much cheaper than a direct drive wheel. If you want to believe I’ve used the ad revenue from PRC to pay for my real world racing adventures, I won’t stop you – it makes for a good story.
We’re now a little over a month into the season, sitting a bit behind schedule thanks to some extremely unfortunate rain-out weekends. After four events, I’ve picked up two heat race wins, finished second in two different feature races, finished third in another, and find myself sitting second in points to my car owner and last year’s track champion on sheer consistency alone. Rookies with no prior racing experience, and surrounded by a field of drivers who have been at this for several years prior, aren’t supposed to be this competitive immediately out of the box. These are drivers from all age groups, posessing multiple years of experience under their belts, and with tons of support from either family members and/or friends, and here I’m just this random guy Justin found playing rFactor 2. On paper, I should have absolutely no idea what I’m doing, and judging by some of the failed attempts at putting prominent sim racers into the real thing, should be throwing up in the pits and a rolling hazard to others. It’s basically the opposite.
Rather than outline the results of every single race in an extremely lengthy blog post, I’d like to take this entry in a much different direction. Obviously, after sim racing for so long, and then getting to drive in a competitive environment away from the keyboard, there’s a couple things I’ve discovered from my travels so far, and I’d like to publicly share them in the hopes that a developer or two can implement my findings into future versions of their software. This post is about to get a whole lot more boring and dull.
There are a diverse array of random sounds that no racing simulators have bothered to model, and they damn well should in order to help drivers to prepare for the real thing. Yes, RaceRoom Racing Experience added in a whole shitload of chassis sounds and miscellaneous cockpit garbage to up the immersion factor, but it’s so much more than that. When I’m pacing, I can pick up bits of the public address system blasting out the track announcer depending on where I am on the track – usually in turn one. If there’s a big moment on the track, someone stuffs it up and nearly loops their car somewhere near the rear of the field, I can hear the audience collectively gasp provided I’m near that section of the facility. If some asshole left a random bolt floating around in my car, not only can I hear it skating around behind me, I can accurately judge how fast its moving from one side of the car to the other. The chassis creaks heard in RaceRoom are indeed present, but they feel like only 40% of the overall auditory experience.
Running over marbles left on the track by Sportsman or Late Model entries… These are not just minor bits banging off the underside of your car, they are fucking loud and a constant distraction. The big chunks sound like gunshots, as if your tire spontaneously blew. If those pieces of rubber are bouncing off your windshield – which they do quite frequently – it’s not an inaudible bug splat, it sounds like someone is pelting the front windshield with tennis balls. All of this little garbage adds up and could do a lot for modern racing sims. Project CARS at least experimented with some of the above effects I’ve mentioned, but a lack of time and resources prevented Slightly Mad Studios from completely fleshing out these immersion effects. I could see drivers getting distracted by this shit in real life, but I found myself thinking “oh wow, this is really cool, racing sims need this” when driving in a pack last night.
And lastly, there’s the contact. We’re not allowed to intentionally dump people, but this is short track racing, and these cars do have fenders. Contact is loud, scary, and disorienting. Even minor door banging, stuff that would be harmless fun under caution at Bristol in iRacing, is loud as hell. It’s so loud, your brain immediately says “don’t do this ever again.” No racing sim has gotten this right as of yet. Shift 2 Unleashed was on the right path with how the helmet cam would aggressively rattle around after contact, but the reality is that the experience is more of an audio overload and a slight head wobble than this super over-the-top camera shake.
Force Feedback Has a Long Way to Go
Your traditional force feedback effects system in any modern racing simulator is designed to condense what a driver feels from the suspension, tires, gravitational forces, and his ass, all through a plastic steering wheel. As a result, certain effects overpower each other, while other effects feel outright unnatural, making the wheel jump around in ways that simply don’t occur in real life. The truth is, while high caliber wheels offer tons of sliders to adjust to your liking, the reason you can never get the wheel to feel 100% right, is because game developers are sending the wrong inputs through the wheel. Instead of focusing on a weird combination of G-Forces, suspension activity, weight transfer, and “canned effects”, the emphasis should be solely placed on tire behavior, because that’s what you feel through a real steering wheel. Early builds of Assetto Corsa in Early Access were quite good in this regard, hence why the game received such high praise from pretty much everyone including myself, but slowly moved away from this type of Force Feedback as the title matured.
During the first three or so laps in any session, the way the steering wheel in my Cavalier feels replicates what you probably feel at all times holding a plastic steering wheel – again, I’ll come out and say Assetto Corsa from the middle of 2014 was extremely accurate in this regard, and all developer teams should seriously study what Kunos were doing back then. However, as the tires heat up on my real car, it’s like a switch has been flicked, and an entirely different force feedback mode has been enabled. I can physically feel the tires deform, flex, shed pieces of themselves, and sustain a dynamic variety of abuse based on my inputs. Depending on the corner, how hard I’m driving, or what line I’m running, it’s a very uncomfortable but extremely informative feeling. It is not a small and insignificant shudder, but an integral piece of feedback.
Right now, racing sims model tire wear as “the car slides around and the steering wheel feels light.” This feels like someone has tried to create plastic Hot Wheels tires that never flex, and instead use complicated algorithms that determine when the tire has grip, and when it does not. A tire is a round piece of rubber that heats up, cools down, deforms, flexes, and sheds pieces of itself inside a competitive environment. Yes, sometimes it does slide around, and in the case of last night’s event, sometimes I’m driving the final five laps with the ass end of the car hanging out, but that’s not all it does.
And that’s just tire wear under normal circumstances. I haven’t even started talking about what happens when you run over a piece of rubber left over by something like a Late Model or Sportsman slick. It’s not uncommon to run the middle groove, pick up a piece of rubber, and develop a fucking massive vibration – one which would frighten most sim racers and convince them they have a serious problem on their hands – when in reality all that happened is you ran over a random piece of rubber on the race track.
I will say though, that catching slides is basically the same in real life as it is in a sim, so developers are making progress when it comes to Force Feedback. As you can see in the shot above, my car owner thought it would be “funny” to dump my ass going into Turn 3 as a form of hazing. I basically did what I’d done a few weeks earlier in RaceRoom Racing Experience to save the car – retaining the position in the meantime.
Shitty Netcode Simulates Momentum Transfer
As I’ve said earlier in the article, we’re not allowed to outright dump people, but this is short track racing after all. Contact is a part of every race, and when we’re not trying to actively move people out of the way, pushing someone to help them get by a slower car is a tactic pretty much everyone uses. On restarts, I’ve become notorious for driving like your stereotypical iRacing asshole and pushing someone to the green flag, but in doing so, unintentionally discovered something hilarious only sim racers will find useful. When you hit someone in an online race, it’s as if there’s a slight delay while the game calculates how much momentum needs to be transferred from your car to the guy you’re hitting. A lot of people actively complain about shitty netcode contact physics, especially in Assetto Corsa, but believe it or not, this delay compensates for the highly complex damage model God has used in our current laws of physics.
In a real car, when you push someone from behind, the front body work partially absorbs the impact of the slower car until the force is applied to the frame of his car, which at this point the force is then completely transferred into the other car, killing your momentum while applying a speed boost to the car in front of you.
Racing simulators don’t model crumple zones, meaning this delay in momentum transfer doesn’t exist – offline, at least. All sim racing cars are made of unobtanium, an infinitely strong substance that begins at the front bumper, and ends at the rear bumper. Sure, you can lose bits and pieces of your vehicle thanks to rudimentary damage calculations, where “If X amount of force is applied to Z component, player is penalized with Y damage to Z component”, but the outright lack of a crumple zone means the slight delay of real car-on-car impact isn’t modeled at all. When you hit an AI car with your front bumper in a sim, you’re hitting them with a brick. The front bumper panel does not compress backwards towards the frame, finally making contact with the frame, and then transfers energy to your intended target.
However, latency in an online race acts as the crumple zone delay during contact. All of those weird incidents where you’ve nailed someone in a corner, and they’ve been affected by it a quarter of a second later, that’s what moving people in real life feels like. Last week I moved a guy up the race track for a heat race win, and I basically had to execute the maneuver at an earlier spot in the corner to accommodate what I could only describe as real life netcode. That slight momentum transfer matters, and if developers are lazy and don’t want to implement simulated crumple zones in their upcoming simulators, they could just not fix the netcode problems.
Managing an Event is Easy
Once you get to the level myself and other extra-terrestrial beings are at in the world of sim racing, managing an event is basically how you win any sort of major online race, and it’s something we outlined how to do efficiently through several chapters in Black Flag: A Crash Course in Sim Racing. You need to know what goals to set for yourself during practice. You need to know what you should be doing during pace laps. You need to be comfortable in traffic. You need to pay attention to your gauges. You need to listen to your spotter and radio feed. You need to consistently glance at the caution lights, flag man, scoreboard, and your mirror. You have to play games and be defensive on the restarts. You need to devise strategies on how to pass other drivers. You need to be confident in your abilities to avoid wrecks. The art of being successful in a virtual environment does not change when your ass is in a real car – racing is racing.
We get two practice sessions in the afternoon, prior to the spectator gates opening. I do not use these for fun, they are almost like work, with the sole goal to diagnose any handling issues. During pace laps, I don’t aimlessly stare into the stands or make obscene gestures to my buddy beside me – I warm my tires and brakes in the same manner as my procedure in Stock Car Extreme. I check my gauges in front of me at a similar rate as I do in some rFactor 2 races on Race2Play. When race control comes over the radio to bitch at someone for unnecessary contact, I’m not startled by the voice from above – half the time, I run online races while casually chatting with bros on Teamspeak. All of the little shit you do as a high level sim racer in any sort of meaningful online race, it all completely transfers over to the real thing, and is 100% applicable. The only difference, is that you’re occasionally looking in different spots. The lap count isn’t planted firmly in the top corner of the screen; I have to take a peek as I’m exiting turn two. What sim racing lacks in immersion or authenticity thanks to the fact that you’re sitting behind a desk holding a toy steering wheel, it makes up for in replicating the procedure and tactics required.
I spent a whole bunch of time on Race2Play from January to May of this year, as well as participated in a league with the guys at Realish Racing, running in a few events every weekend to get my brain thinking in terms of auto racing strategy. As a result, real world events simply aren’t stressful. If I get fucked over on the grid order and start near the rear, I’m not paranoid that I won’t finish well – my brain simply recalls what I did during the reverse grid race at Cascavel in the shitty Brazilian touring cars. If I get shuffled back after a chaotic race start, I can calm myself down as I would on iRacing and dig through the field. Chasing down second place last night, it wasn’t Dawson McGill in the #47 Neon in front of me, I viewed it as Todd Laribee in the K&N cars from iRacing. And when I finally caught up to him, I used a lapped car as a rolling pick to get by him, just as I would on iRacing – a move your average rookie doesn’t know how to pull off. Yes, your body is getting thrown around and subjected to all sorts of G-Forces, and these can scare people who aren’t ready for it, but in terms of raw strategy and event management, sim racing is a bloody good training tool.
Setup Changes are Identical
The isiMotor fanboys are going to jerk each other off over this one.
The first few events of the season went fairly well – I had been exceeding expectations in the standings, routinely busting out quick times in practice, and demonstrating I could handle a car well beyond what my rookie status indicated. For the third race, Justin decided it was time to throw some advanced setup tricks at my car for that extra boost of speed, as aside from getting fucked over on starting spots (we don’t qualify), I was usually right there at the end of the day, and inexperienced drivers don’t normally display this level of competence. The topic of disconnecting the front sway bar eventually came up, an adjustment you can usually only make once on a 2004 Cavalier thanks to how flimsy the sway bar actually is. I was down for it, he was down for it, but in the back of my mind, I was a bit skeptical. On my road course cars in most modern racing sims, I detach the rear sway bar for improved corner exit grip, but in the back of my mind I just didn’t believe removing a fucking part from the car would be safe. I didn’t wanna be that armchair mechanic making setup suggestions based off of what I did in a video game, only to put the car in the wall and embarrass myself. Cause, you know, that would suck.
So I fired up rFactor 2 and put my ass in the Clio Cup at Joesville, a small oval track bundled with the vanilla install that’s just a big longer than where I’ve been racing at on Saturdays. The idea was to drive a few practice laps in an environment that closely resembled the car and track I’d been driving in real life. I ran a whole bunch of laps with the default setup; the quickest time I could achieve was a stout 18.17 after the track rubbered in a bit thanks to rFactor 2’s real road technology. When I couldn’t get any faster, I completely detached the front sway bar in the garage menu to resemble what Justin wanted to do to my real car, and my quickest time dropped over a tenth of a second, to an impressive 18.05 – a pretty big jump in oval racing. The car felt better absolutely everywhere through the corner, as if the front end was much more planted than it had ever been before.
We got to the track a few days later, and I ran the practice session with the front sway bar still attached in its default configuration, busting out a 16.10 – quickest in my practice split. Prior to the heat races, Justin hacked off my front sway bar, and basically said “it’ll feel different, but you’ll figure it out.” I responded by going out and winning both heat races, cranking out a lap time of 16.13 in traffic, and was on pace to sweep the entire event until my soft compound tires gave up with seven laps to go, allowing Justin to get around me for the win. However, upon examining the lap time readouts, I discovered I’d actually ran my quickest time of the weekend – a 16.08 – with the detached sway bar. Just like in my late night rFactor 2 practice sessions with the Clio at Joesville, removing the sway bar from the Cavalier earned me a tenth of a second out on the real track as well. You simply can’t beat that level of accuracy.
As the season progresses, there will most likely be one or two more entries similar to this, outlining some of my observations I’ve made when it comes to comparing virtual racing to the real thing. I don’t plan to use PRC.net as an outlet for my real racing adventures, so if you fancy that sort of coverage, most of my social media accounts are public, and I’ll try to post an increased amount of relevant content on there as well. I’m definitely happy that all of this time invested into wheeling a plastic steering wheel has seemingly paid off, and I now find myself in the thick of a championship battle with nothing but sim racing experience to guide me.