Let’s face it, simulating drift cars and drift racing is well outside Automation’s scope. But it can’t hurt to think about how it might pan out, right?
besides, I love drift, it’s so damn childish
It probably goes without saying that drifting as we know it is the art of deliberately throwing your car sideways while moving under power, and holding a slip angle. More angle and more speed generally means more style points, though how these things are judged varies from series to series. But drifting is not something that any of the cars in Automation do, since they’re instructed to run to the limits of their grip and no more. Similarly, the values for drivability and sportiness in cornering are based on grip, and the wheelspin factor is a penalty.
So, for interest’s sake (and because I want to defy the Automation metric), how does one assess a drift car in Automation? It probably pays to know how one is set up in the first place. For quick, simple reference:
In summary, these characteristics are necessary:
- Powered wheels must be able to easily break traction
- The suspension and chassis must be set up for terminal oversteer
- The suspension and chassis must have as little understeer at any speed in cornering as possible
- The car still needs to have some semblance of control during the drift itself
To facilitate this, and for competition homologation purposes, these rules are generally in place:
- Manual transmission
- Everybody uses the same grade tyre
- No driving aids aside from possibly power steering
- Minimum safety requirement
- No active components in suspension
Also I’ve never heard of active aero being used in a drift GP, but that’s because they tend to be pretty tight on what production cars are allowed in, and that definitely never includes hypercars (although Daigo Saito has a MR converted Liberty Walk Drift Murcielago…). And you’re thrashing these cars like crazy, generally tuning the crap out of the engine, and almost invariably smashing panels off it on most runs. But we’re not here to simulate real life… so much as enhance it. So I figured, I’d like to also explore drift cars that you wouldn’t see in real life, i.e.
- Hypercars and other non-drift car classes e.g. SUVs, trucks, minivans…
- Non FR formats i.e. AWD and MR. Not so sure it’d work with FF because that would involve lift off oversteer which is the opposite of what drift emphasises (unless you’re Jason Plato at Brands Hatch https://www.youtube.com/watch?v=QuTNmVoUR5E)
So here’s what I propose for scoring:
- Breaking traction is important. So the wheelspin penalty will be used to determine how easy it is to do so in your car and therefore just how easily you can initiate and hold a drift using throttle. However, I’m aware the wheelspin calculation only calculates the first couple of gears, AFAIK, so input on how this might be used to calculate higher drift entry speeds would be helpful. Maybe being guided by the accel/grip graph?
- The low speed handling graph will be the most useful for determining how easy it is to steer the car into a drift. I’d recommend the yellow line sticking as close as possible to the red line, and generally achieving an S coefficient as close as possible to 1.00 is the best way to do this. Also note tyre stagger will affect the shape of the curve, so degree of tyre stagger may play a small role
- Maximum cornering G is important because this indicates how fast you can hold a drift around a corner. I’m not yet sure whether I will make this an independent factor, because of later considerations
- Drivability is also important as a counterbalance, from an Automation point of view. I know everything I just mentioned above would kill drivability, but if you maxxed out all of that and ended up with a drivability of 0, well, it’d be really hard to control the car, right? I’m not yet sure whether to simply make the drivability part of the product to determine points, or make it the focus of a random number calculation in success of drift versus crashing out, which might be more realistic, but invoking random number generators makes me uneasy.
- And the unique factor here… track time. I could make like the GRID Drift GP series where your position isn’t so much the winning factor so much as the multiplier factor when making drifts. In this way then you’d have to consider racking up style points versus driving fast so you get a higher multiplier and try to figure out where your opponents in your division have pitched. I know this isn’t really how drift tourneys work in real life but if I don’t involve a track time factor, there may not be a point to even running the cars on a track without a very labour intensive and convoluted system of measures.
So at this stage, drift scoring would involve some kind of product of wheelspin and cornering sportiness coefficient (exponents may be involved), with overall track time being a minor modifier, and drivability being a major modifier.
I’d also love to use different classes of cars, as mentioned, but we’ll go into that more later. First things first, what do you think? If this idea has potential I’ll start illustrating it in a bit more detail, when I get time.
. I think keeping the car still within control is something that is sought after in drifting, too.
