So you've seen Nissan's LMP1 car and you're baffled. How does it work? Why does it look so different to Audi, Porsche and Toyota's Le Mans cars? Why the heck is it front-wheel drive?

Well, Top Gear has got exclusive access to the project, so I've spent the last weekend at a test facility that's so secret I'm not allowed to tell you anything about it other than that I flew into Atlanta, Georgia. The car, you'd imagine, would be equally off limits.

Uh, no. I sat in on team briefings, watched the gearbox being dismantled, talked to the drivers, Olivier Pla and Harry Tincknell, and was basically allowed to poke my nose into everything. This is Nissan's strategy - they want to be entirely open. After all, who else is going to build a front-wheel drive Le Mans car?

So what I really wanted to do was ask Ben Bowlby what on earth he was playing at. Bowlby is the Team Prinicipal and Technical Director and, you might remember, the man responsible for the Deltawing and Nissan Zeod Le Mans projects. He has form in the wackier reaches of sports car racing.

So we stood around the car and he told me about it.

"We could have copied what Audi, Porsche and Toyota are doing, but it would have been difficult to beat them by doing that," Bowlby tells me. "They've got more experience and have put a lot of resources into their cars.

"But if you look at the regulations you'll see that while the aerodynamics at the rear of the cars are very controlled in terms of the wing you can run and the underbody diffuser, they're much more open at the front. There are opportunities in the regulations."

So, the front splitter is basically an inverted wing, channeling huge amounts of air under the car, but diffusing it almost immediately to create downforce, rather than waiting until the back of the car.

The trouble is what to do with all this used air that's still under the front of the car. Well, Bowlby has created two huge air tunnels, which run the length of the car from just after the front splitter all the way through to the back. And when I say huge, I mean huge. Bowlby showed me a picture on his phone of one of the mechanics entirely inserted into one of the tunnels.

So all that air that's used at the front of the car is got rid of cleanly and efficiently and also has the benefit of reducing drag at the back of the car as it's pushed out over the top of the rear diffuser.

The trouble is that if you've generated all this front downforce, you need to have done so for a reason. You might as well put the weighty components up front. And then, since the front axle develops more downforce than the rear, it makes sense if the front wheels are driven.

I know from Bowlby's previous projects - I've been lucky enough to drive both the Deltawing and Zeod, that they work by harmonising the weight distribution, aerodynamic grip and mechanical grip. This one just works completely the other way round from the other two. Spun 180 degrees, if you like.

"The weight distribution is about 65 per cent on the front, 35 on the rear," says Bowlby. "So we have wider front tyres than rears to balance that out, and the downforce is in proportion, too."

With me so far? That, essentially, is the justification for driving the front wheels. I know, I know, like me you're probably still scratching your head, but we're not motorsport engineers so we'll just have to nod and take it at face value.

The packaging under the bonnet is astonishing. The front of the car is so low and yet packaged in there, hidden under a seething mass of cables is a 3.0-litre twin turbo V6. It develops about 550bhp and revs to around 6500rpm. Bowlby's very pleased with it not so much for its power, but its efficiency.

The thermal losses are low and it processes air very cleanly (the exhausts exit through the bonnet. On the one hand, that's very cool, but yes, they have realised it could be distracting for the drivers, especially at night. They're working on it).

But the Le Mans LMP1 regulations demand hybrid technology to achieve an output competitive with Audi, Toyota and Porsche. Not electric, necessarily, but a recovered energy system. You'd assume Nissan would go for electric after its Zeod program, but no.

What they are currently evaluating is a purely mechanical flywheel hybrid system that uses conventional gears and clutches to spin up an 8kg lump of steel and carbon inside a vacuum. So when the driver brakes, an ECU decides whether it should slow the car using the conventional brakes or the inertia of the 8kg flywheel. That's clever.

The testing they're doing out here is the first time they've had the flywheel system working and it's all about making sure the software is working well enough that the driver can't tell the difference and that the flywheel is behaving as it should.

The figures associated with it are hilarious. Running at maximum it can gain 19,000rpm per second, so a braking event of a little over three seconds can spin it up to its 60,000rpm maximum speed. At that speed the outer edge of the flywheel is subject to 47,000g and is spinning at Mach 2. If it wasn't in a vacuum you'd hear a sonic boom. That's right, the GT-R LM could create its own sonic boom. Awesome.

On a more mundane level, the flywheel can send those forces back to the wheels just as quick as it harvested them, at an equivalent of about 700bhp.

These are early testing days, the rear driveshafts haven't been hooked up, so the whole system - internal combustion and flywheel - is driving the fronts alone. That's right, if they let the GT-R LM sing, it's a 1250bhp front-driver. And it might stay that way.

"The plan is to harvest from the fronts and deploy through the rears, but that's not the only way to do it," Bowlby says. "Imagine if we turned up to Le Mans with front drive only..."

Such is the point in testing. All that Nismo is doing here is running straight-line trials. In a few weeks' time they go to Sebring for circuit testing, a session that Audi and Porsche will also be attending. And that'll be fascinating.

Words: Ollie Marriage
Pictures: Greg Pajo