What we've learned from Formula E testing

For the most part, the 2016/17 Formula E season will not be fought out on circuits anything like Donington Park. So drawing conclusions from the final days of testing before the series' Hong Kong opener on October 9 is tricky.

That said, there were insights into the competitive order, what technological changes have occurred between campaigns and the fresh challenges teams are facing.

RENAULT IS THE BENCHMARK

After Renault's Z.E.15 helped Sebastien Buemi to the drivers' title and e.dams to a second straight teams' crown, the onus is on the French manufacturer to pick up where it left off.

Last year Renault pioneered a rear-end architecture that targeted a low centre of gravity. The inverter was moved from atop the battery to lower down, while carbonfibre casing for the gearbox helped reduce weight.

Renault's Vincent Gaillardot neatly sums up its task in developing the Z.E.16 as "optimisation of the powertrain". He says it knew where it needed to improve, with the move to a single gear as opposed to two understood to be the most significant change, and it clearly has.

The Z.E.16 is the pick of the powertrains, ruling the roost when it came to 200kW lap times at Donington. Buemi ended the six days with a new lap record (the first sub-1m29s lap of the track for an FE car), despite the Swiss driver and team co-owner Jean-Paul Driot insisting that the fastest time was not a priority.

"The first year we had all the fastest laps and I didn't win the championship, so I don't really care," says Buemi.

Driot was similarly to the point: "I don't care about being the king of Donington. I want to be the king of Hong Kong.

"Funnily enough we're here to test, and not just the performance over one lap. There are so many parameters to take into account, and you need time to check all of these."

One of the others areas in which the works Renault team was so impressive was the mileage it covered. Not only was it fast, but it also piled up the laps.

It completed more race runs (of a significant distance) than anyone else, culminating in an impressive final tally:

LAPS COMPLETED
1 Renault 572
2 DS Virgin 510
3 Abt Audi Sport 502
4 Mahindra 478
5 Jaguar 464
6 Andretti 441
7 Techeetah 424
8 NextEV 398
9 Dragon/Faraday Future 391
10 Venturi 389

TECHEETAH TURNS HEADS AS CUSTOMER TEAM

Techeetah team principal Mark Preston calls it "an interesting case study". It's going to be more than that: two teams will run the best powertrain on the grid this season.

Renault's supply of the ex-Team Aguri entry was arguably the main talking point of testing, thanks to Jean-Eric Vergne topping three of the six days and only narrowly being beaten to the fastest time of all by Buemi.

Techeetah's Renault deal harks back to early 2016, when then-Aguri boss Preston was sourcing a powertrain supplier after the team barely even considered building its own, having run the series' original powertrain in season two.

Thus the new Chinese-owned entity, which officially took over Aguri's slot at the end of the previous campaign, was left with little choice but Renault - though, had it had a wealth of options, it's difficult to see why it would have chosen anyone else.

The cost of not being a manufacturer meant Techeetah only shook its car down for the first time the week before testing at Donington, and those official tests were important. As Vergne put it, "these guys had never worked on the car before".

No matter. Vergne went quickest on the second day and topped the times on the first two days of the final week as well. The Frenchman looks a man reborn after a trying year with DS Virgin Racing, and in Donington spec Techeetah looked to have got a very quick handle on how to get the most out of its new package.

"It's still a long way to Hong Kong but I'm confident we can achieve great things," says Vergne. "Last year people who were quick at Donington were quick the whole year."

THIS SEASON SHOULD BE ULTRA COMPETITIVE

Donington is often criticised for being unrepresentative, but there are inevitably glimpses of the competitive order.

On the final day of testing, with no more running before the shakedown in Hong Kong early next month, every team launched a late full-power lap with just minutes on the clock in the final session.

The result was eight of the 10 teams dipping under the 1m30s barrier - compared to just one driver, Lucas di Grassi, last season - and 13 drivers being covered by a second.

The two outliers were Jaguar, which was not chasing performance in testing as the incoming manufacturer has so much to learn, and Venturi, which has been hit by reliability issues in the tests so far. And even those two teams were just over a fraction off the pace.

FASTEST LAPS (BY TEAM)
1 Renault e.dams (Sebastien Buemi) 1m28.910s
2 Techeetah (Jean-Eric Vergne) 1m29.109s
3 DS Virgin Racing (Sam Bird) 1m29.205s
4 Faraday Future Dragon Racing (Loic Duval) 1m29.395s
5 Abt Audi Sport (Daniel Abt) 1m29.422s
6 Mahindra (Felix Rosenqvist) 1m29.713s
7 NextEV (Oliver Turvey) 1m29.717s
8 Andretti (Robin Frijns) 1m29.981s
9 Jaguar (Mitch Evans) 1m30.061s
10 Venturi (Stephane Sarrazin) 1m30.121s

So in qualifying simulations at Donington there's little over a second covering the field. Given that the ultimate pace is quite astonishingly a full second faster than this time last year, with no change in the performance parameters of the cars, that is impressive.

Plus, it's fair to assume that the bottom trio of teams will be closer in Hong Kong given that two suffered some unreliability setbacks, and Jaguar has been playing its cards close to its chest.

And once we get to the shorter street circuits that populate the FE calendar, the spread should naturally shrink.

What's more interesting is the long-run data, which pitches four teams into immediate contention at the top. Taking the longest runs from each team, with Faraday Future/Dragon Racing ignored because it's similar in design to Mahindra and focused on shorter runs, reveals the following:

LONG-RUN AVERAGES (LENGTH OF STINT)
1 Techeetah (Vergne) 1m33.4s (9 laps)
2 Abt (di Grassi) 1m33.6s (10 laps)
3 Renault e.dams (Buemi) 1m34.2s (10 laps)
4 DS Virgin (Bird) 1m34.3s (9 laps)
5 Mahindra (Rosenqvist) 1m35.0s (8 laps)
6 Andretti (da Costa) 1m35.0s (6 laps)
7 Jaguar (Evans) 1m35.1s (9 laps)
8 NextEV (Piquet) 1m35.4s (9 laps)
9 Venturi (Sarrazin) 1m35.8s (10 laps)

The main caveat for this data is we don't know the energy levels each car ended with after their respective runs. For example, Buemi's 0.8s-a-lap average deficit to Vergne would easily be explained if he ended the stint with 10% more energy in his battery.

But it's clear that if there is a split in the field, it will be two very competitive halves.

ONE TWIN-MOTOR TEAM HAS STUCK, THE OTHER TWISTED

Gearbox technology has converged for the second season of individual powertrains, which is unsurprising as teams find out what works and what doesn't with this new technology.

Gone are the four-speed 'boxes, with the likes of Renault believed to be using one gear now and two-speeds the norm elsewhere. This is because the teams were still learning what the range of gears would be last season, and improvements on the motor offset the need for changes. Fewer changes mean fewer oscillations, improving braking stability and not disrupting acceleration.

One of the big questions was going to be whether NextEV and DS Virgin dropped their twin-motor solutions from last season, which were heavy and problematic.

NextEV clearly felt it was capable of making it work. Two motors, but repackaged similarly to what DS Virgin had last year.

While it stuck to its guns in terms of working through its own programmes, only venturing out for anything remotely representing a qualifying simulation on the final afternoon, it is clear to see it's ahead of where it was this time a year ago.

In fact the build-up to 2016/17 compared to its disastrous previous campaign has been like "chalk and cheese", according to NextEV president Martin Leach.

Although DS managed the weight issue better than its Chinese-owned rival last season, it is believed to have reverted to one motor.

"Everyone is converging massively on design of the car," says DS Virgin's chief technical officer Sylvain Filippi. "The weight was the issue of our car.

"If we could do that [two motors] without the weight it would be a really good solution. This year we really focused on weight.

"That in a way limits what options you have, but there are quite a few different ways to get the power you need."

COMPLEX NEW STEERING WHEELS MAKE THEIR DEBUT

After a summer spent testing in private with the old steering wheel, the teams were supplied with an upgraded piece of kit for Donington.

In short, the new wheels offer additional functionality - more can be crammed onto the display screens. But with additional functionality comes additional complexity.

Teams have universally found it complicated to get the steering wheels to work properly, to the point where some figures in the paddock are suggesting that one small issue in Hong Kong could cost them an entire practice session.

Once those reliability kinks have been worked out - and almost all of them had by the final day of testing - the steering wheel is a bit of a gateway to enhancing FE's hidden battleground.

Software is free in the championship. While the powertrains are homologated, teams can make small refinements on the electronics side. And given that the nature of this series makes executing the perfect race more valuable than sheer raw speed, having more options available on a race event (thanks to the steering wheel) means fighting with sharper tools.

Filippi says "there's more of everything on the steering wheel - more torque maps, more regen maps". So teams could now have finer differences between each map, allowing them to home in on the perfect set-up for each circuit.

"If," points out Filippi, "you have the resources to do the simulation work.

"It helps react better to race events on strategy. The more you have to react [with] the better. It's nothing ground breaking it's just more work.

"The focus will be even more on the drivers. They will make the difference. Will it become a software war? I don't know..."

In addition to the new steering wheel, the teams are also running new Michelin tyres for this season. They are 5kg lighter a set and offer improved rolling resistance, which aids slightly on efficiency.

Feedback has been mixed - some drivers have complained that the rears are a bit worse, with the added durability of the tyre making it slightly different to the previous version in terms of warm-up and peak grip.

A GLIMPSE INTO THE FUTURE WITH ROBORACE

Like it or not, driverless racing cars are coming.

Roborace's maiden development vehicle for its autonomous race car ran in public for the first time at Donington as part of the first test.

The machine has Roborace's driverless car technology and an electric powertrain fitted to a Ginetta LMP3 chassis.

'DevBot' first appeared on day two of the test and completed a series of stop-start laps with an engineer driving, before the car ran autonomously for almost a full lap.

It stopped at the final corner, allegedly to follow a safety procedure, but reappeared over the remaining test days and by the second week of testing completed a full lap autonomously.

DevBot has previously run only in private, including one appearance behind closed doors at Silverstone. Its performance is not astonishing - there are known to be autonomous vehicles lapping considerably quicker at American facilities, for example.

But what was impressive was how shocked FE paddock insiders were to see it run.

It's fair to say that even those who fully embrace the technological marvel that is a driverless car were incredibly cynical about the likelihood of a Roborace car of any kind hitting the track.

Well, now it has. And in a relatively short space of time too - the concept was only officially announced last November.

History has been made, so what's next? The plan is for Roborace to support FE races, though it is unknown when a prototype of the final design - which will have no cockpit for a human driver to use - will run on track or appear at an FE event.

But it's fair to say that even those who had severe doubts over the concept are now accepting that it's not going to disappear very easily.

IS FE'S FUTURE AWAY FROM DONINGTON?

Given the amount of criticism levelled at Donington by FE teams and drivers when it comes to testing, the news that official pre-season testing is unlikely to return there is no surprise.

The British circuit has acted as the base for the 10 FE teams and the championship's operational headquarters since early 2014.

As part of that agreement it has also hosted collective tests for the first three seasons, with several of those days being made open to the public free of charge.

But as of next summer, it will no longer be mandatory for the teams to be based at the track. And, with several teams heading for the exit immediately, that weakens the case for Donington to host testing.

It is regarded as an unrepresentative track for the street circuit-based championship to test at because of its smooth surface and low kerbs, but as all of the teams have been based there since the series' inception it's been the most logical choice.

The championship is now discussing what other options there are for collective testing - finding somewhere that isn't a permanent race circuit that can fit all the teams, charge all the cars and house all the personnel will not be easy.

Donington will remain FE's logistics hub, and managing director Christopher Tate says he hopes a few of the teams will stick around as well.

Given that the circuit was one of FE's key founding partners, it would be unfortunate if it cannot retain some role in the series' future.