The oil shocks of the 1970s sent carmakers into a panic. What would happen, they pondered, if big oil producers continued to drive up prices by squeezing supply, and governments responded by tightening demand through stricter fuel-economy standards?
The short answer is, a disaster for the auto industry. Back then, the gasoline engine wasn't nearly as efficient as it's become in the 40 years since. Engineers would have to find an alternative that provided sufficient power and superior fuel economy.
But they had to act fast. By the late '70s, gasoline in the U.S. had rocketed past a dollar a gallon—a huge increase for the time—and was steaming toward two bucks. There was no relief in sight.
Electric cars, powered by heavy lead-acid batteries, were unsuitable for anything but lightweight, short-distance commuter cars. Downsizing all cars to increase fuel efficiency would only play half-way. The American market still demanded large interstate cruisers and pickup trucks—for work and pleasure. The only alternative that made any sense was the diesel engine.
American companies needed a diesel solution that would work for a Silverado or an Oldsmobile Station wagon. Mercedes-Benz, at the time, built the world's best diesel engines for large passenger cars, but the Stuttgarters were not about to share.
General Motors sent its engineers on a mission worthy of the Apollo 13 rescue crew: to refit its ubiquitous small-block V8 with diesel-ready heads. It was not an ideal solution: Diesel engines' combustion comes from intense compression of fuel and air, not by way of a spark plug, and even the stout V8 blocks had trouble withstanding the added pressure.
The resulting Oldsmobile 5.7-liter diesel was deemed appropriate on the test bench, but in the real world—as installed in Cadillacs, Buicks and other large cars and mid-sized trucks—it was failure-prone. Some say it single-handedly ruined the perception of diesel engines in the U.S. for the next three decades.
Still, GM was the first mover, and captured 60 percent of the total U.S. passenger vehicle diesel market, which by 1981 totaled 310,000 units.
As the 1980s unfolded, Ford was running behind. It needed a diesel solution that could work for its Lincoln Continental Sedan and Mark VII coupe, but refitting one of its V8s was not an option. What would it do?
The answer came from Munich.
Back in the mid-70s, BMW engineers had developed the M21 diesel engine to compete with Mercedes-Benz and Peugeot, the two automakers that dominated the European diesel market.
They started with the standard BMW M20 straight-six engine, but installed new valves, pistons and crankshaft, and planted a turbocharger on top. The 2.4-liter engine produced 115 horsepower and 154 lb-ft of torque, taking the BMW 524td from zero to 60 in just under 13 seconds, which, while slow as a wheelbarrow in mud, was still quicker than the Mercedes-Benz 300D.
It was this motor, built in a joint operation by BMW and builder-for-hire Magna Steyr in Austria, that Ford plugged into the Continental and Mark VII. The result was a slow but relatively sturdy package, that delivered somewhere around 30 miles per gallon in the luxe Lincolns. Lincoln advertised the diesel-powered Mark VII LSC, in close proximity to a Mercedes-Benz SEC, as being available with a "European designed 2.4-liter Turbo Diesel." Thanks, Bimmer.
Ultimately, bad timing killed the sad American diesel luxury car. It was a mercy killing. Gasoline engines would soon gain enough efficiency to fend off a supply crunch, just as fuel prices fell precipitously throughout the rest of the '80s. That left the the diesel boom-and-bust as a weird footnote in the American auto industry of the 20th century.
Postscript: BMW's M21 also found its way into a weird motorhome called the Vixen. Google it.
The Silver Arrows of the past will once again lap the racetracks of Europe, thanks to the upcoming Mercedes-Benz Classic Trackdays events. Former Mercedes-Benz factory DTM drivers will join customers to relive the vintage racing experience.
Classic touring car fans will recognize the three venues from the golden era of German touring car racing: Hockenheim, Zolder, and Oschersleben. These historic tracks will host the Classic Trackdays, which will take place in March, August, and September, respectively. Mercedes-Benz Classic has even recreated an authentic copy of a 1990s Group A 190E 2.5-16 Evolution II (W201), one of the most successful and highly regarded touring cars of its era. Former DTM drivers will pilot the car around the same racetracks where the 190E famously traded paint with the BMW E30 M3 touring car rivals.
Track action will include open track time and a Club Challenge, where participants navigate through courses of light barriers in their personal vintage vehicles. Drivers and their guests will be given VIP treatment with catering and hospitality while surrounding themselves in the world of classic motorsport, and the limited field of 45 vehicles per event will include some of Mercedes-Benz’s finest motorsport icons.
With racing revival events like the Rolex Monterey Motorsports Reunion, Goodwood Festival of Speed, and now the Mercedes-Benz Classic Trackdays, fans of yesterday’s racing series are spoiled for choice when it comes to past eras of motorsport.
China may roll back proposed production quotas for more new electric vehicles sales, Reuters reports, after Beijing got pushback from the automotive industry. The original draft of the proposal—which came out in September—mandates that eight percent of new car sales be from plug-in hybrids or full EVs in 2018, ten percent in 2019, and 12 percent in 2020. Apparently, the rollback would only delay 2018's proposed sales mandate by one year.
The argument the automakers are relying on is not a new one by any means—they claim these sales targets are too high and "could hurt [their] interests," Reutersreports. But considering that EVs and plug-in hybrids only accounted for 1.8 percent of new car sales last year, a jump by 6.2 percent is significant.
An Jin, chairman of Anhui Jianghuai Automobile Group (JAC Motor), said, "Whether the whole market can hit this quota by 2018 depends a lot on the strength of government policy. If it's strong then we should be able to surpass the targets. An added, "If you consider China's infrastructure and the transformation of China's auto sector, then perhaps the pace will have to slow."
According to two executives who are intimate with the proposed draft changes, the government may try and cut the quota by two percent each year or "push back each target by a year." Either way, since the proposal is still in draft form, revisions can still be made.
For the price of a boring new car, why not buy a vintage Ferrari?
Did you know the average transaction price of a brand new car in the US is $33,652? Call it $35,000 with applicable taxes. That’ll get you a nicely-equipped Hyundai Genesis, Chevy Malibu or Nissan Maxima four-door sedan or even a stripper 3-Series BMW. But everybody’s got one of those, and they won’t turn heads or cause a stir when you roll up to the valet.
So think out of the box. Why not buy a used Ferrari for that same $35 grand?
Now we’re talking.
OK, you can’t get a classic Ferrari V-12 berlinetta for even close to $35K, but there are vintage Ferrari two-seaters and 2+2’s, with V-8’s and V-12’s, that you can snatch up for about as much as you’d pay for a new car that starts depreciating the minute you shake the salesman’s hand.
And it’s a Ferrari, so one of these bargains could even appreciate over time.
We combed the exotic car sites, and with the help of price guides from Hagerty’s and Cavallino magazines, we can help you find a much more exciting ride for your money.
The Ferraris we recommend, for the most part, are 2+2’s, so there’s actually room for a couple of passengers and/or enough luggage for a sexy weekend getaway.
Ferrari’s 1960-to-1963 250GTE 2+2 was the marque’s first real volume passenger model, but you can’t get one of those today for buppkes, because they share nearly the same driveline as a multi-million dollar 250GTO. But in 1973, Ferrari replaced the racy, mid-engine 246GT “Dino” with a sharp little 2+2 that previewed the GTB-to-come’s lusty 3-liter, 4-cam V-8. Then the magicians in Maranello, knowing that some clients wanted a front-engine GT car for everyday driving, introduced the 365GT 2+2.
And it gets even better. There really are affordable Ferrari’s right up to the 1990’s. So follow along while we take a quick trip through the byzantine world of used Ferrari’s, complete with some of the challenges of owning a bargain Italian stallion.
Dino 308 GT4: 1974-1980
The Dino 308 GT4 bowed at the 1973 Paris Salon. Replacing the curvaceous Dino 246GT, this angular little coupe was designed by Carrozzeria Bertone, not Pininfarina. The original 246 Dino’s 2.4-liter V-6 had been replaced with a 3-liter, 4-cam V-8 packing four Weber carburetors. It developed 205-bhp at a screaming 7700-rpm. It was really a 2+2 (with tiny rear seats), but that was never part of its official name, nor was there a Ferrari badge anywhere, at first. But by 1976, Ferrari owned up and all 308 GT4’s sported prancing horse badges. The 308 GT4 is fun to drive; its 5-speed shifts crisply and when you nail it, you get all those wonderful Ferrari sounds. Those tiny back seats are best used for luggage, as there’s no rear legroom. 308 GT4’s are still under the radar pricewise. If you can find a USA-legal Euro-spec version, (and quite a few were imported) you’ll get 240-to-255-bhp.
208 GT4: 1975-1980
Here’s a tip to get an even cheaper GT4. Ferrari made 840 208 GT4’s with a smaller bore 2-liter, 153-bhp version of the 308’s 3-liter V-8. This tamer, but visually nearly identical 2.0 version came about because Italian tax laws heavily taxed 3-liter cars. Since you couldn’t use the 308’s 155-mph top speed anyway, even on the autostrada let alone on Highway 101, Ferrari S.p.A. offered a milder 208 for the home market, and they are stone bargains – especially because you can bore out the 4-cam V-8 to 3-liter specs and have all the goodies. 208 GT4’s aren’t common, so check the Ferrari Market Letter classifieds for one of these. And don’t tell the Italian IRS!
365GT4 2+2: 1972-1976
Think you can’t get a V-12 Ferrari for $35 grand? Think again. The 365 GT4 2+2 shares the sporty 1971-1972 365 GTC/4’s six-carb 4.4-liter V-12 with six side-draft Webers, but a classic C/4 starts at $250K, and you can double that for a great one. So consider the 365 GT4 2+2 instead. Introduced in Paris in 1972, it’s 7.5-inches shorter than the 365 GT 2+2 “Queen Mother,” it replaced, but with a 2-inch longer wheelbase and more contemporary Pininfarina styling. This lovely coupe was the first of a series that includes the later 400GT and 400i, through 1984. With 320-bhp, knock-off alloys and fully independent suspension, they’re fast, elegant Grand Tourers. Borrani wire wheels were still an option. One caveat: they’re much quieter than their sportier brethren, but you can fix that fast with a Borla or Tubi stainless exhaust update.
400GT/400 Automatic/412: 1976-1989
When Ferrari updated the 365 GT4 to the 400 GT at the Paris Salon in 1976, buyers could opt for a 5-speed or (gasp!) a GM-supplied THM400 4-speed Hydra-Matic. It was a sign of the times. Well-heeled clienti wanted a Ferrari, but they didn’t want to shift for themselves. Not surprisingly, automatics soon outsold the 5-speed sticks, so be prepared to pay a $5 grand premium if you want a rare manual. Displacement rose to 4.8-liters and output was an impressive 340-bhp. Bolt-on 5-star alloy wheels replaced the original model’s knock-offs. Borrani wires were no longer offered. Sadly, to meet US emission regs, the Bosch K-Jetronic, fuel injected 400i dropped to 306-311-bhp, then 315-bhp by the end of 1982. The 412’s became lusty 5-liter cars with 340-bhp once again. They are even nicer-looking, thanks to body-colored bumpers, and a higher rear deck with a discrete spoiler. Forget the Ferrari dealer’s expensive service department: your local AAMCO can probably fix that GM-sourced tranny.
208 GTB/GTS: 1980-1982
It’s nearly impossible to find an affordable 308 GTB, let alone a 308 GTS on our $35K budget, unless you encounter a rusty rat or a wreck. But the Italian market, Euro-spec 208 GTB, if you can find one that was legally imported, is worth considering. Like the 208 GT4’s, 208’s are rare in GTB/GTS guise. Ferrari built only 160 GTB’s and 140 GTS’s from 1980-to-1982, but a few came over, so hunt around. Like the 208GT4 2+2’s their 121-cid V-8’s developed just 153-bhp, so they look fast, but alas, they’re not!
208 Turbo (GTB and GTS): 1982-1985
But here’s the solution: When the tax-relief special normally-aspirated 208’s for Italy were deemed too slow, Ferrari offered a turbocharged version of the coupe and the spider, from 1982 (GTB) and 1983 (GTS) until 1985. The turbo’s output is 217-bhp @ 7000 rpm; that’s a healthy 64-bhp more than the 2-liter, normally-aspirated model.
Again, these are thin on the ground over here, but you may be able to find one.
Mondial 8, QV, 3.2, t: 1980-1993
Ferrari built 3,571 Mondial 8’s in several variations, because there were clients who wanted a more spacious, marginally less sporty 2+2 coupe or convertible with a Ferrari prancing horse badge. The wheelbase was 4-inches longer than the 308 GT4’s, so there’s really room for two passengers. Among enthusiasts, the Mondial is kind of the Rodney Dangerfield of Ferrari’s, but most civilians haven’t a clue. They simply see a handsome red coupe or convertible with a Ferrari badge and they’re impressed. Just remember, it costs just as much to repair the transversely-mounted, 4-cam V-8 in a Mondial 8 as it does in a 328 Berlinetta. Mondial 8’s, built from ’81-to-’82, and the QV (Quatrovalvole) coupes and true convertibles, offered from ’83-to-’85 , are well within our budget. Mondial 3.2’s (with 260-bhp) and later Mondial t’s in average shape make the $35K cut, but the better examples can go for as much as $10-to-$15K more. You have to do your due diligence, and be sure to inspect all existing service records. A belt replacement on one of these puppies will seriously blow the budget.
Tips on buying and servicing a used Ferrari:
OK, so you’re ready to find the affordable Ferrari of your dreams. Ferrari’s aren’t generally listed in local newspaper classifieds or “pennysavers,” but you will find lots of Ferrari ads in Hemmings Motor News, the Ferrari Market Letter, and the top British car mags like Octane, Classic & Sportscar and Classic Cars, and at the broad appeal auctions like Auctions America, Mecum, and Russo & Steele.
No matter how tempting the price, (unless you’re a skilled mechanic, with a stash of metric tools), if the owner/dealer doesn’t have the car’s service history, (or at least the last few years worth), simply move on. Assuming the service records check out, be sure to see what next major service is needed and make that price part of your negotiations. Ferrari’s were subject to rust, so a careful inspection of all the body panels, the chassis, etc., is mandatory. Inspect for any evidence of accident repairs. Ensure everything works, right down to the windshield wipers and back-up lights. Check the exhaust system for leaks. Nothing on a Ferrari, no matter how trivial, is cheap to do.
Here’s the dirty little secret: Ferrari was very cavalier about service. They figured if owners could afford the car, they could afford to have it maintained. On many models, you have to remove the engine for belt and timing chain service. Spare parts aren’t cheap. Neither are tires. Michelin TRX’s on later cars may not be made much longer. Many Ferrari’s were carefully garaged and maintained, but some of the less expensive examples may have been abused or suffered flood or accident damage. If there’s a CarFax report available, by all means get it. All the customary used car buying rules apply here. Drive the car for at least 20 minutes. Watch the water temperature, Check for unusual noises or vibrations. Try not to fall in love until you’re sure it’s a worthy example. If you’re importing a car, be sure you know the requisite DOT/EPA rules (there are too many to detail here).
So why are we encouraging you?
Ferrari’s are fun, exciting, exhilarating. Chicks dig ‘em, at least until they’re trying to decide if you’re a responsible person. Owning a Ferrari taps into 70 years of wonderful prancing horse history, on the track and on the road. Unless your Ferrari’s in the shop, you’ll smile every time you see it. So go for it—but be careful and purchase wisely. Life is not a dress rehearsal. We don’t get to do this twice.
Bargain Ferrari’s by the numbers….
Years made Model name Number Produced Price Range
1972-1976 365 GT4 2+2 521 $30K-$90K
1974-1980 308 GT4 2+2 2826 $30K-$35K
1975-1980 208 GT4 2+2 840 $30K-$33K
1976-1979 400 GT Automatic 502 $30K-$60K
1979-1984 400i Automatic 1308 $35K-$60K
1980-1982 Mondial 8 703 $25K-$30K
1980-1982 208 GTB 160 $25K-$30K
1980-1982 208 GTS 140 $30K-$35K
1982-1985 208 (GTB) Turbo 437 $23K-$35K
1983-1985 208 (GTS) Turbo 250 $25K-$35K
1982-1985 Mondial Coupe QV 1145 $25K-$40K
1983-1985 Mondial Cabriolet QV 629 $27K-$45K
1985-1989 Mondial 3.2 Coupe 987 $30K-$40K
1985-1989 Mondial 3.2 Cabriolet 810 $35K-$42K
1985-1989 412 576 $45K-$90K
1989-1993 Modial t Coupe 858 $35K-$55K
1989-1993 Modial t Cabriolet 1017 $38K-$55K
(Source: Cavallino Magazine, October 2016 [Keith Bluemel/Cavallino])
Mazda's CX-5 crossover is new for this model year, and to all of our pleasure, you get more standard features for not much more money. The manufacturer announced that the front-wheel drive Sport model can be had for a shade under $25,000 with upper-level models like the Grand Touring clocking in at $33,465, making it an affordable choice in several trims.
First introduced at last year's Los Angeles Auto Show, the 2017 CX-5 gains new styling and engine choices. A more angular appearance helps to sharpen up the crossover's handsome chin while still maintaining a cheap price point throughout the range. A 2.5L I4 SkyActiv engine will be standard in each model with a 2.2L turbo diesel option coming later, each being paired to a six-speed automatic as Mazda has ditched the manual transmission for 2017.
With the Sport model, you get value. You may not get the frills of fancier CX-5's, but there are plenty of features that come as standard. All wheel drive is available, bumping up costs to $31,000. The upper end of the Sport quickly transitions into the model's mid-range Touring trim.
Mazda promises a lot with the Touring in ways of creature comforts. It provides luxuries like leatherette covered seats with heated fronts, dual zone climate control, blind spot monitoring, reclining rear seats, and LED headlights, all for $26,855. If you opt for Mazda's total safety package, you also gain radar cruise control to finish things out. The automaker also offers a preferred equipment package, a bundle that includes a power liftgate, power moonroof, navigation, and a 10-speaker Bose audio system.
Then, if you really your Mazda, the you'll want the Grand Touring. The top-of-the-line CX-5 once again comes equipped with front wheel drive and all wheel drive availability, an option that'll cost you $3,130. You'll get every option that comes standard in the Touring model along with adaptive headlamps, LED foglights, leather upholstery, and 19-inch wheels. The Premium Package throws in a heads up display, windshield wiper de-icer, and heated steering wheel.
Mazda is yet to announce the release date of the 2017 CX-5. We anticipate its arrival within the next few months, so keep an eye on your local showrooms.
But this is, at best, only partially true. While a number of automakers have engineered vehicles that can pilot themselves with an ability unfathomable even a decade ago, after months of interviews with the people shaping the self-driving car industry it's clear that our autonomous future—the one where you take a nap as your vehicle whisks you to your destination in comfort and safety—is not in any real sense here now, nor around the corner, but likely decades away. All claims to the contrary are either based on misunderstanding or are intentionally misleading.
That's not to say that future isn't coming, eventually. Aside from nearly every notable automotive manufacturer, major tech players like Apple, Google parent company Alphabet, and Uber are currently staking out their respective turf in the ascendant autonomous ecosystem. But to understand where we're going, it's helpful to take a clear-eyed look at where we are right now, ignoring the hype.
Don't Believe the Clickbait
To illustrate the issues of hyperbole, let's backtrack to another headline: "10 Million Self-Driving Cars Will be on the Road by 2020", from Business Insider (June, 2016). A quick scan reveals their definition of "self-driving car" to be incredibly broad: "We define the self-driving car as any car with features that allow it to accelerate, brake, and steer a car's course with limited or no driver interaction." Adaptive cruise control, which covers the acceleration and braking aspects of that definition, has been around since 1998; add lane-keep assist, which debuted back in 2003 and can control the vehicle's steering to keep it from leaving its lane of travel, and you have Business Insider's characterization of an "autonomous vehicle"—i.e., cars like the Mercedes-Benz S-Class and Tesla Model S and a half-dozen other models already on the road. This is hardly the customer fantasy of "Netflix and chill" in the backseat, and given the average creation cycle for a new vehicle is anywhere from three to five years, it seems unlikely we're about to experience a wave of fully-autonomous vehicles by the time the next Presidential election roles around.
An eager marketplace tends to trample or dismiss such discrepancies in favor of the shiny, best-case-scenario version. We've seen it before: we're promised a Jetsons-like ideal of a robot butler—the reality is a Roomba. So, what is the current reality of the self-driving car?
The Limitations of Current Technology
A vehicle's ability to "see" and translate its surroundings comes from onboard sensors. The three major sensor systems in use, alone or in combination, by vehicles currently on the road are: 1.) cameras; 2.) radar; and 3.) Light Detection and Ranging (LIDAR), currently a very expensive system that measures range via pulsed laser, and can be found atop Google’s latest self-driving minivan and Uber’smodified self-driving Volvo XC90s, among others. Google's 64-channel LIDAR system is good to 120 meters, creates a 360-degree image, has a 26.9-degree field-of-view, and can take more than two million readings per second.
"The camera is very good at providing a huge amount of information," says John Dolan, a principal systems scientist at Carnegie Mellon University’s Robotics Institute. "But interpreting that information accurately is difficult because of the lighting issue," he says, referring to the loss of image quality that occurs in situations like direct sunlight, poor or extreme contrast, or fog.
Lasers like LIDAR aren't disrupted by lighting issues, and are "very good at giving you shape information without too much difficulty in terms of the processing," Dolan says. "But it gets confused easily by bad weather." Radar is not confused by weather, "but it doesn't give as much shape as a LIDAR [system]—it gives, basically, just range or distance, and the rate at which the distance is changing, or the velocity of the vehicle."
There is currently no standard for the type of equipment used in semi-autonomous systems. Tesla's Autopilot notably foregoes LIDAR, with CEO Elon Musk stating he's "not a big fan" of the technology due in part to its complexity and expense (though another of Musk's companies, SpaceX, uses LIDAR on its Dragon reusable rocket). In May of 2016, a Tesla owner and avid fan of the brand named Joshua Brown was killed when a semitrailer truck pulled out in front of his Model S sedan on a Florida highway. The Autopilot system, engaged at the time, did not recognize the white trailer and failed to apply the brakes, as did Brown. A six-month investigation into the incident by the National Highway Traffic Safety Administration (NHTSA), which took into account changes that Tesla made to Autopilot after the crash, including more aggressive alerts when drivers remove their hands from the wheel and system disengagement when those warnings are ignored, cleared the company of a "safety-related defect trend,” though The New York Times noted that "some experts speculate that a LIDAR-driven car might have avoided this fatal crash."
(It should be noted that the same NHTSA report declared that "data show that the Tesla vehicles crash rate dropped by almost 40 percent after Autosteer installation" and that Autopilot does not engender "unreasonable risks due to owner abuse that is reasonably foreseeable (i.e., ordinary abuse)," while Musk, during a press conference last year, said that "Autopilot accidents are far more likely for expert users" than novices. More on all that in a bit.)
Unforeseen incidents aside, even predictable weather conditions remain a massive hurdle. Heavy rain alone can shut down an autonomous system, and forget hands-free driving on a lonely snow-covered road. According to Liam Paull, a research scientist at MIT’s Toyota-funded Computer Science and Artificial Intelligence Lab (CSAIL), snow "pretty much hoses both camera and laser, because now everything's just white, you can't see the road lines." And, according to Paull, laser doesn't bounce well off the snow, creating "all sorts of garbage" in the feedback.
As a result of such issues, most major manufacturers regulate when and where autonomous features are available. Eric Coelingh, technical director of active safety functions at Volvo, said the company will start by sending engineers to personally verify the roads on which full autonomy will be enabled on its upcoming production cars.
"We have to test that it works on a particular road in particular weather conditions, and when we know that it's working [during] nice weather in Gothenburg, it doesn't mean that it works with nice weather in New York, or with bad weather in Los Angeles," Coelingh says. "We don't have the data on that."
But even when Volvo does have the relevant data, that doesn't mean the autonomous capabilities function without restriction.
"When the circumstances are out of this [verified] scope because the road conditions are completely different, or the weather conditions are completely different, then we cannot make this autonomy feature available to our customers," Coelingh says.
So, despite the widespread use of the term "self-driving car," what's actually being described is a vehicle with limited semi-autonomous capabilities only available under certain conditions. This is the reality now, and will be for the foreseeable future—and for the foreseeable future, these early-stage autonomous experiments will share the road with traditional vehicles piloted solely by humans. This could be a dangerous combination.
The Space Between
The NHTSA first issued a “Preliminary Statement of Policy Concerning Automated Vehicles” in 2013. It laid out five tiers of autonomous operation, from no autonomy at Level 0 to an occupied or unoccupied car that could handle every operation on its own at Level 4. Last year, the agency adopted a new classification system outlined by the Society of Automotive Engineers, explained in the 30-page document "Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles" that defines six levels of autonomy. Level 0 is intermittent warning systems like blind-spot detection; Level 1 encompasses features that can monitor the environment and alter steering or acceleration and braking, like parking assist, which only controls steering, or adaptive cruise control (ACC) that only adjusts speed; Level 2 includes systems that simultaneously steer and change speed, like ACC with lane-centering, and traffic jam assist that maintains space in traffic and navigates shallow bends. Under all of these level definitions, the driver is still charged with monitoring the environment.
At Level 3, an "Automated Driving System" can take over all driving functions and total monitoring of the environment—the caveat being that the driver is expected to be ready to take over if the vehicle strays off-course, or finds itself in a situation it can’t handle. According to the SAE paper, at Level 3 "an ADS is capable of continuing to perform the dynamic driving task for at least several seconds after providing the [driver] with a request to intervene." So, while the vehicle should be able to fully monitor its environment, the driver must also be ready to take over in an emergency.
That point at which a self-driving system requires a human to retake the controls to avoid an incident is termed a "disengagement." The California Department of Motor Vehicles requires any company testing autonomous vehicles in the state to log every disengagement that occurs; the most recent report from the California Department of Motor Vehicles shows commendable year-over-year improvement between 2015 and 2016. Waymo, the self-driving car project from Alphabet, led the rankings with just 0.2 disengagements per thousand miles driven autonomously in 2016—a four-fold improvement over the previous year even as total miles logged increased by 50 percent.
But although the cars are improving, none of the experts with whom we spoke believe Level 3 autonomy is safe—and that the issue has largely to do with people. A self-driving car operating in autonomous mode can create in drivers a false sense of security and a susceptibility to distraction, making an emergency disengagement more dangerous because the point at which the driver needs to retake control of the vehicle is also a point at which he is poorly equipped to do so. The specification says, in effect, "Your car will drive for you, but you need to watch it while it does, just in case.” Studies have shown humans aren’t good at continuous monitoring without personal involvement, and even Ford engineers tasked with monitoring the brand's self-driving cars from behind the wheel doze off at such a high rate that the company recently announced it would skip Level 3 autonomy altogether. It's an unavoidable fact that when told that their car will perform a task they aren’t interested in, such as driving, humans will do other things—like watch a Harry Potter DVD.
Level 4 is the first phase at which a vehicle will never need driver input, no matter the situation. If a Level 4 car gets in trouble and the driver doesn’t take control, the car is programmed to get itself out of danger—by pulling itself off the road, for instance. (This is one of example of why Volvo sends human engineers to verify roads approved for a future, Level 4-autonomous Volvo: a road without a safe extraction point may not be fit for hands-free driving.) Currently, there are no Level 4 autonomous vehicles for sale, though automakers and headlines continue to insinuate or outright declare they'll be here in three years' time, by 2020:
But what if customers want Level 4 and an automaker only has Level 2? That's when the advertising creatives step in to try to make the latter sound as much like the former as possible. Twice last year, regulators and consumer watchdogs went after automakers over potentially confusing marketing of autonomous functionality.
In March of 2016, Mercedes-Benz released a 30-second commercial called “The Future”, touting its 2017 E-Class sedan. Over scenes of the futuristic Mercedes F-105, a purportedly fully-autonomous but nonetheless concept-only vehicle, the narrator ("Mad Men" star Jon Hamm) asks, “Is the world truly ready for a vehicle that can drive itself? An autonomous, thinking automobile that protects those inside and outside?” At the moment the narrator answers his own question with, "Ready or not, the future is here," the decidedly non-fully-autonomous Mercedes-Benz E-Class rolls into frame. The commercial didn’t explicitly state equivalent capability between the F-105 "that can drive itself" and the "self-braking, self-correcting, self-parking" E-Class, but watchdog groups insisted that equivalence was too strongly suggested.
On July 27, consumer advocates sent a letter to the Federal Trade Commission asking the FTC to "scrutinize" the commercial. On July 28, Mercedes dispatched troops to defend the 30-second spot, with a spokeswoman saying the ad contained a disclaimer about the E-Class's true abilities. That disclaimer, in fine print and appearing for six seconds, reads: "Vehicle cannot drive itself, but has autonomous driving features. System will remind the driver frequently to keep hands on the steering wheel." Much easier to catch was the company's print advertisement timed to run in magazines around the same time as "The Future" ad appeared on television and the internet—a glossy, full-page shot of the E-Class with the text "Introducing a self-driving car from a very self-driven company.”
"Given the claim that consumers could confuse the autonomous driving capability of the F015 concept car with the driver assistance systems of our new E-Class in our ad 'The Future,' Mercedes-Benz USA has decided to take this ad out of the E-Class campaign rotation,” the company said. A version of “The Future” uploaded to the Mercedes-Benz USA YouTube channel in October features a less-suggestive voiceover.
And then there was Tesla, whose situation was, as usual with the Silicon Valley company, more complicated. Tesla controversially named their driving assistance system "Autopilot," after the systems that appear in airplanes, and Tesla's web site notes the technology "classifies [Autopilot] as a Level 2 automated system by the National Highway Transportation Safety Administration (NHTSA)", while in the same paragraph calling it "an advanced driver assistance system" that is "designed as a hands-on experience." As Tesla CEO Elon Musk has repeatedly explained, a plane's autopilot flies under the constant supervision of at least one pilot ready to intervene immediately, and at no time to do cockpit personnel watch movies or go to sleep on the flight deck. Knowing all of that, the name "Autopilot” makes sense (although it should also be noted that studies by NASA and others show that pilots also have a hard time paying attention to their autopilots).
The problem is that the name piggybacks on a very specific definition of "autopilot," one that makes sense at 38,000 feet but is less obvious to non-pilots (i.e. nearly all of the global car-buying public). If someone at sea level said about a machine, "I left it on autopilot," and then went on to explain that she spent two hours monitoring said machine to make sure it was operating correctly, a reasonable response would be, "that's a bad autopilot." That sort of confusion is, in part, why in October 2016 German regulators sent a letter to every German Tesla owner stating that Autopilot “requires the driver's unrestricted attention at all times.” That same month, the California Department of Motor Vehicles proposed a regulation to forbid Tesla from using the terms “autopilot,” “self-driving,” and “automated” in advertising materials. (The proposal never left the draft stage.)
There is likewise an issue with the term "in beta," which Tesla has repeatedly used to describe Autopilot. While the term is defined by Merriam-Webster (and common usage) as "a stage of development in which a product is nearly complete but not yet ready for release," Tesla has always insisted that Autopilot is a fully-vetted system ready for primetime, and that the term, as Musk explained at a press conference last year, is used "to reduce people's complacency in using [Autopilot]" and "not because it's 'beta' in any normal sense of the word." Which begs the question: when it comes to such a high-stakes feature as Autopilot, wouldn't using any definition other than the "normal sense of the word" lead to confusion as to the system's actual operating state? And building on that, a larger question taking into account both Musk's assertion that Autopilot accidents are much more likely for expert users than novice ones and the preponderance of anecdotal evidence showing Autopilot users operating the system incorrectlyor irresponsibly and counter to Tesla's own definition of "intended use" as compared to systems from other manufacturers: is it possible that the term "beta" confuses some Tesla owners into thinking that Autopilot is still in a test phase, and an enthusiastic and risk-taking subset of those owners (which would include people like Josh Brown) are deliberately pushing the boundaries of the system in the name of analysis?
We posed that question to the company. A Tesla spokesperson issued the following response:
Autopilot's 'beta' label is intended to reduce complacency, and Tesla has taken many steps to prevent a driver from misusing Autopilot, including blocking a driver from using the system after repeated warnings are ignored.
While that answer clearly skirts the question, the truth is that regulating advertising language or policing automaker jargon can't fully control the message. There is another, arguably larger problem.
The Media Hype Machine Is Putting Lives in Danger
Consider the potential for mixed messages in this Bloomberg interview with Betty Liu:
In the video, Musk explains that the driver is ultimately responsible for what happens behind the wheel, even when Autopilot’s engaged. In the same interview—but in a video clip posted separately—Musk rides with Liu in a Model X, activates Autopilot, puts his hands in the air, and says, "It’s on full autopilot right now, I’m not touching anything, no hands, no feet, nothing.” Musk then continues the interview while the Model X autonomously drives, navigates bends on the Tesla campus, and stops behind another vehicle.
Musk was simply demonstrating his new technology at the request of a media outlet. But consider how that media is presented, and consumed: at the time of this writing, the video about driver obligations has 166,798 views; the video of a self-driving car autonomously ferrying several passengers, posted on the same day, has 1,027,268.
Tesla owners as well as members of the media posting hands-free Autopilot experiences have become a YouTube staple. Like this one. And this one from CNET that begins with the line, “You don’t need to put your hands on the wheel to drive a Tesla.” And this one from Inside Edition about a grandmother “Behind the Wheel of Self-Driving Tesla.” And this one, with the line, “It was clear we were in good hands with the P90D, but with all the new free time the Autopilot afforded us, what were we going to do with it?” And this one from Josh Brown a month before he died, showing his Tesla avoiding a collision while merging onto a highway. In the video's description, Brown wrote: “I actually wasn't watching that direction and Tessy (the name of my car) was on duty with autopilot engaged.” What was in "that direction" Brown wasn’t watching? The highway onto which he was merging.
Bryant Walker Smith, an assistant professor in the University of South Carolina School of Law and an affiliate scholar at the Center for Internet and Society at Stanford Law School, says "the levels of automation were described as a promise by the automaker to the user: 'We are promising that our system will do this under these conditions.’ So, part of this will be clearly communicating those capabilities and those limitations to the user, not just once, not just in an owner’s manual, but in real time, as the systems are operating.”
This is where Tesla's "disruption" cuts both ways. It’s inconceivable that the CEO of a major manufacturer would tweet that one of its marquee convenience systems, installed on a production car on sale to the public, is still in beta, as Musk has done, or that an OEM exec would repeatedly say the law is just going to have to catch up to the company’s innovations. To be sure, the risk-averse legacy car manufacturers certainly still get ahead of regulations and operate beyond them at times. But, for numerous reasons, they don’t crow about it, because it tends to be bad for business.
For now, Tesla’s upgraded its current Autopilot with new safety restrictions. For the next version of Autopilot, Tesla has added cameras, updated the sensors, and improved the radar processing (while still holding back on LIDAR). But the rise of autonomous systems from a growing number of manufacturers creates an uneasy situation on the road, according to MIT's Liam Paull.
“There’s plenty of people out there who are happy to take the risk for the thrill of having an autonomous vehicle,” Paull says. “In general, I'm OK with people accepting risks and taking risks, but there's plenty of other people on the road who have not agreed to these kinds of things.
"It's difficult, because on the one hand you need a good regulatory environment to push the science forward, but we all have a responsibility to do things properly.”
When Silicon Valley Met the Auto Industry
Manufacturers, tech companies, and consumer groups broadly agree that the federal government needs to set the tone for coordination. In an attempt to answer that need, the National Highway Traffic Safety Administration last September released its 116-page Federal Automated Vehicles Policy. The guidelines for “highly automated vehicles” (HAVs)—the kind with driving controls that humans can still operate when necessary—assigned responsibility for regulating the hardware and software to the federal government; created a 15-point safety assessment for design, development, and testing of self-driving vehicles; and drafted a Model State Policy that “outlines State roles in regulating HAVs, and lays out model procedures and requirements for State Laws governing HAVs.”
Stakeholders were pleased with this long-awaited first step. Nevertheless, among the more than 1,100 public comments submitted on the policy, those same stakeholders listed numerous complaints about issues the NHTSA both did and did not address.
Establishing a national framework pleased the states, mostly; the National Conference of State Legislatures regretted that the NHTSA hadn’t asked the NSCL for input before drafting the Model State Policy, and noted possible inconsistencies and ambiguities in the division of federal and state responsibilities. Apple wanted exemptions for “limited and controlled testing on public roads,” plus clearer rules on the gathering and use of driver data. Lyft didn’t believe the NHTSA went far enough to prevent mismatched laws from state to state that would hinder streamlined development nationwide. The Property Casualty Insurers Association of America wanted a more clear understanding of liability and state oversight in the event of autonomous crashes.
Outside the scope of the policy paper, others have suggested how the NHTSA might handle regulating an industry suddenly filled with new and niche players working beyond traditional vetting and recall procedures originally designed for major automotive manufacturers. Among his numerous publications, Walker Smith wrote a 100-page paper titled "Lawyers and Engineers Should Speak the Same Robot Language" that addressed this very issue. When asked about potential pitfalls in this early development phase, Smith said that "one of the questions we’ll really struggle with is how to deal with a less top-down industry, because "major automakers, suppliers, Silicon Valley companies that do this research will have strong reputational [and] financial interests to act deliberately."
"There will be some smaller startups that may or may not have those same interests," Smith says. "And that could be concerning for the federal government and for state governments with respect to road safety.”
Carnegie Mellon scientist Dolan concurred. "I think that the current and maybe uneasy marriage between a Silicon Valley mentality and a Detroit (i.e.) more cautious traditional automaker mentality has had some fruitful results," he notes, adding that "the historical traditionalism of the automakers has been broken through to some extent by the freewheeling attitude of Silicon Valley, and I think as a result we've made more progress more quickly than we otherwise would have."
"However, I think that the reliability and testing practices of the automakers need to be harnessed or somehow introduced into the process to a greater extent, so that we don't have any unfortunate problems."
A series of high-profile crashes, either fatal or injurious, that mar the public's faith in the technology would no doubt constitute an "unfortunate problem." And because autonomy is new, exciting, and puts lives at stake, we judge the technology on a higher standard—one that turns every mundane fender bender into a high-profile event.
"What we really need are federal standards," she says. "They're trying to jump the gun and come out with these cars without federal standards in place, just basic standards that say, for instance, that [autonomous cars] can't be hacked."
Proper regulations would give the government a bright-line rule for issuing recalls to protect public safety. "All they've done so far is draft guidelines," Shahan said, "and that's not adequate to protect consumers. You have to actually have a standard that [automakers] need to meet. And occasionally they fail to meet those standards and then [the authorities] issue a safety recall. Without a standard, it’s unclear what [it would] take to get a safety recall issued, and what will be considered an unreasonable risk to safety.”
The NHTSA doesn’t have the money, staffing, engineering capacity, or tech know-how to codify a set of all-encompassing laws in advance. So, as Walker Smith told Wired last May, our autonomous future could be "driven by what actually gets to market." The rest of us will simply be along for the ride.
When Will Robots Win the World Cup?
What, exactly, will that ride look like? In other words: what do those currently guiding the industry envision as a likely roll-out for autonomous vehicles? The people we spoke to mostly agree that, in terms of the vehicle, the hardware is ready, it’s the software that needs finishing. Self-driving cars don't need to learn how to navigate roads as much as they have to learn how to navigate humans. Despite massive processing ability, an autonomous car still doesn’t know how to handle what MIT’s Paull called “corner cases,” or “the part of the space that you don't get a lot of data about because [contact] happens so infrequently, so it's very hard for a robot to learn what to do in those situations.”
Corner cases can include the wide variety of situations that require a sort of best-case judgment call, such as navigating a residential street with legal two-way traffic but a single usable lane. Human drivers use waving and finger pointing to communicate with one another and establish order; computers don’t speak that language yet. Nissan, BMW, Ford, Volvo, and Toyota have pledged to sell a Level 4 autonomous car by 2021, and the standard five-year automobile development cycle means they’re working on those cars right now. Considering the challenge of teaching the hardware our human ways, computers are unlikely to have their own wave-and-point capability by 2021, either.
Ford has already suggested that its coming Level 4 car (scheduled for 2021, naturally) will operate “in a geo-fenced area that is very heavily 3D mapped.”
According to Volvo engineer Coelingh, “the best self-driving cars very soon will be better than the worst drivers we have around—the inattentive drivers or the drunk drivers. But before [these cars] can beat the really attentive drivers, that will take time.”
That means, at the very least, a lengthy and phased roll-out of autonomous capabilities—or, as Paull puts it, "having a car that can [drive itself] in progressively more and more scenarios." This, "rather than a switch where there's a car that can do it all," is the likely scenario. "And the way that starts is with highway driving,” he says.
Of the type of Level 4 autonomy that relieves a driver of the need to monitor the environment, Walker Smith says the vehicle needs "two of the three [things]: slow speeds, simple environments, or supervised operations."
"I could see a company like Tesla having a system that technically does not expect that the user will pay attention on freeways," Smith says. "I could see in the next year or two a company deploying a low-speed shuttle under relatively simple conditions where the people are just passengers.”
Dolan, of Carnegie Mellon, says current systems are "largely able" right now to deal with routine highway driving and simple city roads. About Level 4 autonomy, even on the highway, he was more circumspect when taking corner-case scenarios into account.
"If you stick to a given route at low-traffic times of the day, it may well be that an autonomous car can run 95 percent of the time without human intervention," he says. "But if you enumerate all the different things that can go wrong, or [are] difficult in urban driving, autonomous cars today will not be able to handle 95 percent of them.
"How long is it going to take for robots [to be able] to do that? It's similar to asking how long is it going to be until we have a robot team that can win the World Cup."
Volvo demonstrates one of the approaches we can expect in the interim. This year, the Swedish carmaker will lease 100 development cars to select customers in Gothenburg for its "Drive Me" pilot program. One goal of Drive Me is to "research how ordinary people want to use this technology" in preparation for the autonomous car Volvo expects to sell at dealerships in 2020.
That 2020 production car will supposedly take the carmaker from its current Level 2 assistance features straight to Level 4 autonomy. Volvo's Coelingh told us, "That car will have three different modes: you can drive it manually, and you will have all the latest and greatest safety systems that will help you to avoid collisions; you can drive in assisted modes like ACC and Pilot Assist, and these two modes will be available, in practice, anywhere at any point in time; the fully autonomous mode, the Level 4 mode, will also be in that car, but you will only have it available on certain roads—roads that fulfill a number of preconditions.
"We have designed the technology for your daily commute, so we will select roads that a lot of people use to get into a big city or out of the big city—the bigger freeways, or roads where there's a lot of congestion. And then the car will tell you whether full or autonomy is available. But I do not expect that by 2020 it will be available at any point in time, on any road. That will take much more time.”
Volvo has already said that when it makes Level 4 autonomy available, it will accept liability for accidents that occur during the proper use of the autonomous system.
The Drive Me initiative will work with the city of Gothenburg to explore open and covered parking solutions for large numbers of autonomous vehicles. "What's even more interesting [than the technology itself] is that [it] is an enabler for making the traffic system much safer than it is today," Coelingh says. "There's lot of focus on technology, but there's actually very good reasons to look at the bigger picture and try to understand how this will impact society. I think it’s more interesting that it can really change the transportation system as we know it.”
Toyota will take a different approach, practically backing its way into Level 4 autonomy with the help of an army of scientists working at two well-funded research centers at MIT and Stanford. "The core idea here is that, instead of building an autonomous car, you build a very advanced safety system that's basically an autonomous car that sits in the background and makes sure a human doesn't do anything wrong," Paull says. "That’s got a couple of advantages. One, you never have to go through Level 3. The other is that, if your system decides it's not sure what's going on, it can always do nothing. The hope is that you should do no harm; you should have a system that, worst case, is as good as a human driver, and the best case is a car that's incapable of crashing.”
Toyota’s CSAIL institutes are one year into Toyota’s five-year funding commitment. The Japanese automaker says it expected to have an offering with autonomous capability on-sale by—you guessed it—2020.
Miles and Decades to Go
This article touches on some of the hurdles facing full autonomy, but certainly nowhere near everything that governments, manufacturers, and the public will need to sort out in the meantime. For instance, upgrading the national infrastructure so self-driving cars will have the lane markers they need. Establishing liability protocol—if a "driver" isn't driving her car, is the manufacturer at fault if that vehicle crashes, or maybe the software supplier?—is itself a massive undertaking, and will likely tie up a lot of time in various courts across the country. There’s the seemingly trivial but potentially crippling matter of human bullying; in other words, how to make an autonomous car safe enough to "do no harm” yet aggressive enough that it doesn’t continually give way to humans who figure out that a self-driving car, unlike a person, is always programmed to stop for pedestrians? (Plus, some of those pedestrians could be among the thousands of human drivers predicted to be out of work thanks to self-driving vehicles.)
What about the possibility of hacking self-driving car? Despite a lot of hand-wringing, the full scope of possibilities still seems unclear.
It will take years, money, miles, legislation, and judges to figure it all out—and that's just for the problems we can foresee. Even the bullish Elon Musk has said Autopilot needs to log a billion test miles before it gets out of beta; according to some reports, Tesla would be only just north of 200 million active Autopilot miles by now.
So, what about those autonomous Jetsons dreams? We’ll get there, eventually, but it will be slow and it will come piecemeal. And that’s assuming the technology doesn’t suffer any major missteps. As to the question of a realistic timeline for you to be able to get in your car and ride to your favorite restaurant without touching the wheel or putting down your book, Paull says we might get "a car that can do 95 percent of the driving tasks without any oversight from a human within ten years. But that last five percent is going to be very, very tough." Volvo’s Coelingh says, "If you're talking about anywhere, in any weather conditions, I think it’s longer than 20 years.” Gil Pratt, CEO of the Toyota Research Institute that oversees the two CSAIL outposts, said at this year’s Consumer Electronics Show we’ll need "decades to have a significant portion of US cars operate at Level 4 autonomy or higher."
So not even the boffins know how it’s all going to play out, or when. Walker Smith says that "the things that we expect are going to take far longer, and it’s the things we don't expect that are going to be right around the corner." Regarding that first wave of autonomous cars being promised by the manufacturers in three to four years, Smith notes that "you might not be able to take this car from your house 1,000 miles to a mountain range, but you might be able to take it on your freeway, and you might be able to take a shuttle around downtown LA. Or you could just have flying cars and drones that we all find imminently more interesting than self-driving cars. Who knows?”
Who knows, indeed? Anyone who claims otherwise has a robot butler to sell you.
One could argue that car shows are events specifically organized for the purpose of showing off, but that means parking your car and popping the hood—not launching your vehicle into traffic and causing accidents. Unfortunately, some people never learn.
Videos uploaded to the internet over the past weekend show an orange second-generation Chevy Camaro pulling out from a parking lot with the driver losing control almost immediately. First, the Camaro aims itself toward a curb, then slaps back around and t-bones an oncoming Chevy Impala. According to a Facebook commenter and the YouTube videos, the incident occurred outside a Cars & Coffee event in Oklahoma City on Saturday.
From the looks of things in the clip, the impact was relatively low speed, but it was also probably completely avoidable had the driver pulled out of the lot in a more calm fashion.
By now we've already seen the new Panamera variants scheduled to be arriving at the Geneva show, both the new wagon-shaped Sport Turismo, and the insane hybrid-powered turbocharged super sedan Panamera Turbo S E-Hybrid. Both of those cars are extremely exciting for Porsche fanatics like ourselves, but this morning Porsche teased a third debutante model for Geneva, one that nobody expected we'd see, at least so soon. Buried in an otherwise unexciting press release, Porsche put just one sentence that really perked up our spirits. "Power as passion is a key feature of the third world premiere in the form of a motorsport-derived two-seater, which creates an intensive connection between everyday driving and the racetrack." Obviously, that's no Panamera variant.
The wording of that sentence is so important, Porsche doesn't use words by accident. Many people will jump immediately to the conclusion that this sentence confirms their Cayman GT4 RS suspicions, but I think it might be something else, as an RS upgraded Cayman would likely not be described as "an intensive connection between everyday driving and the racetrack". Even the most masochistic among us would have a hard time driving any Porsche RS model on a daily basis.
Furthermore, 'motorsport-derived' can mean a lot of things in Porsche-land these days, as they currently race a total of five distinctly different race cars, with three different variants of 911 [the mid-engine RSR, the FIA GT3-spec GT3 R, and the one-make-series GT3 Cup], as well as the 919 Hybrid LMP1, and the Cayman GT4 MR. Which of these could this new model be 'derived' from?
The further clue here is the fact that it's a two-seat car. Of course, this could be a two-seat 911 variant in the same vein as a GT3/GT2/911R. In fact, there are reports of a GT2 having been seen testing recently. Likewise, the alleged GT4 RS would fit this part of the description, but it must be something else entirely. As it happens, I think this might be a new Porsche we've never seen before. There have been rumors for ages that Porsche is developing a new mid-engine platform to take the fight directly to McLaren, Ferrari, and Lamborghini. It's all speculation at this point, but a 919-inspired motorsport-derived hybrid super-sports car would fit in the image that Porsche is currently trying to build for themselves.
What do you think this new daily-useable motorsport-derived two-seater car is? Let us know in the comments.
The wide body Porsche 911s from Rauh-Welt Begriff have been a point of controversy for some time. There are those who are disgusted by the act of chopping up a classic air-cooled Porsche, and there are others that believe 911s are not only meant to be driven, but also personalized. Still, even within the camp of RWB enthusiasts, there are still heated debates regarding what an RWB Porsche should, and shouldn't, look like.
Which brings us to RWB Medusa:
It's a classic RWB wide body Guards Red 993, with front canards, rear fender vents, and an oversized GT wing. The body kit looks even more purposeful with Lexan windows, complete with sliders on the door, and vents over the rear quarter. A matching Guards Red colored roll cage and GT3 bucket seats further show the RWB Medusa's motorsport inspirations. Finally, a set of copper Volks TE37 wheels finishes off a proper execution.
But then we get to the vinyl work, namely the giant Supreme decal across the side of the car, and that's where things start to get a little touchy. For those that aren't familiar with the brand, Supreme is a boutique skateboard apparel store, which first opened its doors in lower Manhattan in 1994. What was once a store that catered to a small niche of New York skaters is now a global icon, known for its unique designs and its extremely limited runs. Even though Supreme has intentionally kept its physical growth slow in order to retain its brand identity, it still managed to spawn a following of fanatic collectors and admirers that, over the years, would either wait in front of the store for countless hours or drop an obscene amount of coin for "deadstock" on eBay in order to set themselves apart.
People who aren't familiar with Supreme will never understand what all the hype is about. At the same time, veterans of the Supreme movement may feel betrayed as they witness a new generation of cultists dilute the brand's sense of exclusivity and authenticity it once possessed.
But doesn't that sound familiar? Replace Supreme sweaters with RWB Porsches and you'll recognize many similarities between both the brand as well as the people that relate to it.
The mainstream couldn't care less about the RWB Medusa, although a select few may commend the owner's decision of the Supreme vinyl and recognize its place as a symbol of status and individuality. However, RWB 'purists' (is that a thing?) may find Medusa's new Supreme livery a big slap in the face to Akira Nakai's bespoke Japanese aesthetic.
Last week, Streetwear blog High Snobiety shared a new video that profiled JM Yang, the owner of RWB Medusa, where he inevitably talked about the decision to add oversized Supreme decals on his RWB. In response to the naysayers, JM shared something that Nakai once told him: "Feedback is good, whether it's positive or negative."
Would you rock this Porsche 993 RWB Medusa? Check out High Snobiety's video below:
Mad Max: Fury Road proved to be a smash hit in the box office and a great homage to the original 1979 Mad Max when it was released in theaters in 2015. Successfully winning Best Production Design at the 2016 Academy Awards, Fury Road was filled to the gills with highly modified and attentively detailed machines that proved to gain traction among the car communities and custom builders. And now, we have video of what one of those roadgoing creatures looks like when placed on an actual highway outside of Hollywood.
The video below takes an obsession to Mad Max: Fury Road to a whole new level. In the clip, you see a full-sized replica of the "War Rig" rolling down the highway. Although there are minor inaccuracies welded onto the trailer, the general idea and styling is identical to the original build. This begs the question, who is the talented welder behind this build?
With a project this size, it is safe to assume that this trailer has made its way around local meets or at the very least the forums. If you know this talented gentleman (or lady), please comment below.