Richard D A Jones' blog on the relative development costs of the McLaren F1 and Bugatti Veyron made me think about what those costs indicated.
A quick dig reveals the different design goals:
Developed by McLaren Cars and designer Gordon Murray, the motivation behind the F1 was to build the best street car ever, regardless of cost. The design goal was to create the highest power–to–weight ratio possible in a user–friendly, roadworthy machine. To this end, the McLaren F1 was the first road car to use a complete carbon fiber chassis and body with lightweight composites and exotic metals throughout. A magnesium–cast BMW engine mated to a very lightweight (and very expensive) transverse gearbox further helped weight savings. (the engine was a version of the BMW850csi powerplant)
Whilst the Veyron:
The goal was create the world's fastest production vehicle.
They said that the car was going to have 1000hp, and that it would reach 250mph and still be driveable. They said it would be aerodynamic, AND look good.
They developed a new W16 engine, gearbox wth 10-20 years life etc,etc, Prototypes did 50,000 mile endurance tests to ensure that reliability was good enough.
So the design objectives of the F1 are more modest and are perhaps can be classed as "only" a substantial innovation. I wonder if the F1 can be driven by an ordinary person for thousands of trouble free miles..... but that wasn't the objective.
The Veyron's objective on top speed and power and reliability mean that it is a much more radical innovation challenge.
Jeremy Clarkson said in his review of the Bugatti...
"At 200mph you can feel the front of the car getting light as it starts to lift. As a result you start to lose your steering, so you aren’t even able to steer round whatever it is you can’t see because of the vibrations. Make no mistake, 200mph is at the limit of what man can do right now. Which is why the new Bugatti Veyron is worthy of some industrial strength genuflection. Because it can do 252mph. And that’s just mad — 252mph means that in straight and level flight this car is as near as makes no difference as fast as a Hawker Hurricane.
You might point out at this juncture that the McLaren F1 could top 240mph, but at that speed it was pretty much out of control. And anyway it really isn’t in the same league as the Bugatti. In a drag race you could let the McLaren get to 120mph before setting off in the Veyron. And you’d still get to 200mph first. The Bugatti is way, way faster than anything else the roads have seen."
But it is all nonsense really. Both cars are corporate excercises. The smaller McLaren produced a roadcar that could race competitively with its peers. The Veyron is a corporate strategic exercise of iconic proportions (See Maslow). As a technology platform project it may be that the engine, transmission, etc. will find themselves in other prestige brands, and top end mainstream vehicles.
Maybe the best lesson is dramatically ambitious goals can drive timescales and cost incredibly high. The two are related of course as developer's salaries are a major determinant of cost and longer timescales often mean additional resources are sucked in to address the technical and time issues (usually the timescale obstinately does not shorten!.). I remember being told that development timescales were determined by the experiences and collaborative skills of the team. The formula was:
1. If the team had never ever tackled this sort of project then Actual Time (AT) = Estimated Time x pi x pi, approx = 10 times ET
2. if the team leader has done a similar project but the team hasn't
then AT= ET x pi, approx 3 times ET
3. If all the team have worked together on a similar project then
AT= ET x pi/2, approx 40% greater than estimate.
Small wonder that Jeremy Clarkson wrote
'"God, it was hard,” said one of the engineers I know vaguely. “The gearbox in an F1 car only has to last a few hours. Volkswagen wanted the Veyron’s to last 10 or 20 years. And remember, the Bugatti is a damn sight more powerful than any F1 car.”
The result, a seven-speed double-clutch flappy paddle affair, took a team of 50 engineers five years to perfect.'
At the end of the day the point Richard Jones was making a very valid point "When companies have a big objective they need to invest early in planning and studies. Correcting a mistake on paper is several orders of magnitude less expensive than a correction during development and is far cheaper than a recall." which I will blog about shortly.