The deep and lasting significance of the McLaren MP4/4’s forgotten older brother

Eagle-eyed readers will have noticed that there is something not quite right about the images with this article. The car looks like an MP4/3, the McLaren in which reigning double Formula 1 world champion Alain Prost scored three wins on his way to fourth in the 1987 standings.

But it has ‘Honda’ adorning the engine cover and nose, whereas the team’s power that season came from the TAG Porsche programme that had helped McLaren to three drivers’ titles and two constructors’ crowns across 1984-86. And what about those snorkels on the sidepods? 

This is one of two MP4/3Bs built to help ease Ron Dennis’s McLaren into a new era. They were never raced but provided an important testbed, not just for the Honda engine that would go on to power the legendary MP4/4 to 15 wins from 16 races in 1988, but also for a relationship between the Woking squad and the Japanese automotive giant. 

Honda had taken both the drivers’ and constructors’ titles in 1987, with its RA167E V6 strapped into the back of the Williams FW11B that won nine races in the hands of Nelson Piquet and Nigel Mansell. But Honda was allegedly not pleased with how Williams had managed the rivalry between its drivers and was also a big fan of Ayrton Senna, twice a winner with the Lotus-Honda 99T.

At Monza for September’s Italian Grand Prix, Dennis was able to announce that, for 1988, McLaren would have Honda power and the services of Senna, creating a superteam with Prost that would set new standards. And the design team was only given the heads-up shortly beforehand. 

“We’d been running the TAG turbo for a few years, so going to a fully funded supplier was a step forward,” confirms McLaren designer Neil Oatley. 

“By then the TAG had served its purpose,” adds Indy Lall, who was tasked with setting up what would become an extensive test team. “We knew the new chassis wasn’t going to be ready so we wondered what could we do to help Honda get used to the new regulations. It seemed ideal to have a testbed.” 

Those new regulations were a cut in the fuel allowance from 195 to 150 litres per race, plus a reduction in the boost from 4.0 bar to 2.5 bar. Normally aspirated cars were allowed unlimited fuel and could run 40kg lighter than the turbo’s 540kg minimum. The idea was to make cars with normally aspirated 3.5-litre engines, which were to become the norm in F1 from 1989, more competitive against the 1.5-litre turbos, but it didn’t turn out that way… 

Prost’s 1987 Portugal victory in the ‘standard’ MP4/3 set a new F1 record for world championship GP wins

Photo by: LAT Images via Getty Images

“There was lots of analysis and Honda felt it was still an advantage to have a turbo engine, even with the new restrictions,” recalls project leader Matthew Jeffreys, who adds that the point of the MP4/3B was “mainly to get experience working with Honda, develop the relationship with the engineers, and to get a car up and running with a Honda engine as quickly as possible”.  

“It had to be quite a quick turnaround,” adds Oatley. “The design office was small and we were tight on time, but we wanted something to run the engine in. We needed to build a good reputation with Honda; we needed to be at the top of our game.” 

The answer was to fit a modified RA167E, run to 2.5 bar and with revised turbos, sump height/oil scavenge and combustion chambers, into an existing MP4/3 chassis. The engine would then be replaced by the RA168E when it was ready – the new powerplant would not run on the dyno until January 1988.

The car’s purpose would be to prove such things as packaging, cooling, fuel consumption and pop‑off‑valve behaviour before the MP4/4 was available. This included the fitting of Honda’s preferred option for oil/water heat exchangers, not used on the TAG-engined cars. One car, MP4/3B-6 pictured here, was assembled around an unused monocoque, while MP4/3B-2 was converted from the car damaged when Stefan Johansson famously hit a deer during practice at the Osterreichring.  

“It had to be quite a quick turnaround. The design office was small and we were tight on time, but we wanted something to run the engine in. We needed to build a good reputation with Honda; we needed to be at the top of our game” Neil Oatley

Perhaps surprisingly, those involved remember the conversion being relatively easy. “The Honda was a tighter package, but it didn’t look any different except for the snorkels, which were critical for Honda, and the installation was quite seamless,” says Lall. 

“It was quite straightforward, just a question of engine mounts and the piping,” reckons Neil Trundle, then a travelling fabricator before being made chief mechanic for 1988. 

One of the benefits of the RA168E package, which arrived shortly before the race season began, was how low it could be run. The MP4/4 would get a special three-shaft gearbox to benefit from this, but that wasn’t an option for the MP4/3B, which had a high gearbox input shaft. 

“One of the main differences was the crankshaft centre line height was an inch less,” recalls Oatley. “For ease of getting the car running when the new RA168E was ready, it was mounted an inch higher than it could have been. We had a special modified two-shaft gearbox casing to mate with the higher engine position – so the crankshaft was at ‘normal’ car height.” 

RA168E engine was a significant step from its predecessor, aimed at suiting the restrictive 1988 rules

Photo by: LAT Images via Getty Images

With either engine the car was compromised but outright performance wasn’t the goal – getting up and running was. Prost gave the hybrid a shakedown at Silverstone in December 1987 before heading to Estoril, which was so badly hit by rain that the team stayed an extra day. A programme over the winter was then conducted in Japan, usually with newly recruited test driver Emanuele Pirro at the wheel. 

“I remember it with pain because it was quite a short chassis and I’m quite tall – it wasn’t comfortable at all, but I always liked technology and development, so the programme suited me,” says the Italian.

“What was required was not speed but accuracy and analysis. A pace quick enough to fully understand the car. Because the car was very small in the foot area, the first time out when I started to push at Suzuka I got to the hairpin and missed the brake pedal completely, I went straight on and crashed. It was not a bad crash, more painful to my pride, but the British are very phlegmatic and it wasn’t a big problem.” 

It immediately became apparent that McLaren was now working with a giant and that the stakes had been upped considerably. Oatley describes the Porsche/Bosch involvement as relatively low key. It had undoubtedly been effective, but Honda was “full-on”: “The financial constraints weren’t that intense,” which is surely engineering speak for ‘they had a big budget’! 

“Most of the frontrunning teams were at the first Estoril test and the engine ran quite reliably,” continues Oatley. “Their electronics guys were good and we had much bigger back-up than we’d experienced with TAG.” 

“TAG were super, super people and we did well, but Honda was a completely different powerhouse,” adds Lall. “It quickly became evident they were super-clever people but not necessarily practical. They just needed some direction and, once they accepted we were all on the same page, they were fully committed. The evolution of any concept was phenomenal, second to none.” 

“Working with Honda was amazing, they had multiple engineers assigned to every problem; we had multiple problems for each engineer!” says Jeffreys. “The time difference helped a lot, too: you could ask a question by fax just before you left for home in the evening and the answer would be waiting for you when you walked in the next morning.” 

But it took a little time. Pirro recalls being taken aback initially: “It was quite strange with the Honda people at first. I’d not met them until the day before the first test. There were all these computers that looked like Cape Canaveral and I was asking questions, but they were very reluctant to share information, even the maximum RPM for certain corners. I thought, ‘Oh dear’, but slowly but surely we established a really good relationship.” 

Opinions are divided on how competitive the MP4/3B could have been – but there was never an intention to race it

Photo by: Mouse Motors

Pirro formed a strong partnership with Tim Wright, who had scored title success with Prost and engineered Johansson during 1987. Wright constantly flew back and forth between Europe and Japan for the tests, spending periods at Honda’s Wako R&D centre. 

“I spent a lot of time with Honda sorting the oil tank but I don’t remember many other issues,” explains Wright, who worked on the oil problem using both the dyno and the MP4/3B. “The biggest development was things like the throttle scrolls, and Honda wanted a lot of sensors. It was early days for a lot of things like telemetry and sensors, so there was a lot of work around that. All that knowledge was sent back to Woking so they could make any changes that were necessary. And it let Honda get on top of how the engine performed. 

“Pirro’s input was so important – he was a great guy to work with and his feedback was always good. He was brilliant and it was great fun setting it up and getting to know Honda.” 

Ideas began flowing between the two parties. The operation would eventually extend to a completely independent team operating out of Wako in Japan, with its own crew and trucks, plus a two-car test team in Europe and the three-car race operation. It’s hard to imagine today in the era of tight testing restrictions and cost caps, but McLaren and Honda were pushing F1 into a new realm. 

“My task at the end of 1987 was to set up the test team from nothing and it was full-on, trying to recruit and get the equipment. Japan was so alien but their warmth and openness made it a lot easier” Indy Lall 

“My task at the end of 1987 was to set up the test team from nothing and it was full-on, trying to recruit and get the equipment,” says Lall. “Japan was so alien but their warmth and openness made it a lot easier.” 

Performance on-track was also encouraging. Despite the regulations knocking around 200bhp off the engine’s output, Wright remembers Prost being “quite surprised, fairly impressed with the engine” from the beginning with the MP4/3B. 

It was also the first McLaren that Senna drove, aside from his 1983 test in an MP4/1C. The Brazilian “wasn’t keen on the chassis side – he didn’t like the handling very much”, reckons Wright – though, given Senna was coming from Lotus and its active suspension, it is unclear how much of that was due to the basic MP4/3 design and how much due to the compromises of the Honda test car.  

At a Rio test, Senna even handed the car back to Prost to confirm set-up direction, though Oatley does not believe changes were very significant. “The set-up of the car wasn’t wildly different – with 200bhp less we had to run lower downforce settings and that meant the suspension could be softer,” he explains. 

Big announcement: superteam of Senna, Prost, McLaren and Honda is made public at the 1987 Italian GP

Photo by: LAT Images via Getty Images

The programme racked up a lot of running. “The Honda was very reliable – we weren’t throwing engines away,” says Lall. “Changes were due to mileage or because of developments. The car gave the drivers an idea on the engine and it was primarily for Honda to see how they could develop for 1988 – a testbed for efficiency, reliability and power.” 

It also gave McLaren an early taste of Senna’s abilities. Wright recalls a Bullet train ride back with his then-new driver after one Suzuka test: “He went through a circuit map and talked about corner entry, mid-corner and exit, what he was feeling, for every corner. It was amazing he was able to recall that a few hours after testing, but that was Ayrton.” 

At the same time, Honda pressed on evolving the powerplant. Michio Kawamoto, chief designer on both the 1987 and 1988 F1 engines, remembers the extensive process and explains that little was carried over from the RA167E to the RA168E. 

“Honda’s development plan for the next season began in early summer,” says Kawamoto. “The basic specification and drawing was fixed in December, and assembly of the new model was at the end of January to deliver to McLaren in February. 

“The engine for the Silverstone test in December was based on an RA167E with modified parts and the engine could be updated day by day. The engine for the Imola test [in March 1988] was the RA168E, the same as for April’s Brazilian GP.  The cylinder block and the cylinder head were new because the crankshaft and combustion shape were different, and the piston and connecting rod also changed. The bearing size of the crankshaft was also decreased due to low friction, so there were almost no common parts between RA167E and RA168E.” 

The progress on the engine side only added to the pressure on McLaren to deliver a car worthy of it, with Gordon Murray, Steve Nichols and the rest of the Woking team working to get the car ready ahead of the opening round in Rio on 3 April. “The confidence was there, but the question mark was the chassis,” concedes Lall. “We were late.”  

If that seems a little surprising, it’s worth remembering that McLaren’s previous title winner, the MP4/2, had been designed by John Barnard, who had left during 1986. The MP4/3 had been a winner but not a champion, and there were high expectations from Honda. 

But, remarkably given the number of changes going on, the MP4/3B was competitive. At the Rio pre-season test in March, Prost recorded sixth fastest-time with 1m29.86s, while Ferrari’s Gerhard Berger set the pace with 1m28.51s. At the Imola test later in the month, when most of the teams were present, Prost beat Berger, 1m29.30s versus 1m29.90s.  

Senna drives the MP4/3B during pre-season Rio test; Autosport’s preview bigged-up Ferrari but also made Prost title favourite!

Photo by: Ercole Colombo / Studio Colombo / Getty Images

So, would the MP4/3B have been competitive if it had raced? Enthusiasm for the question varies. “We were there or thereabouts, though it’s difficult to know what the other teams were doing,” says Oatley, while Lall is less convinced: “We arrived at the final Imola test and it wasn’t competitive. I’m not sure what the fuel capacity was so we might not have been compliant! The weight distribution also wasn’t right because it was designed for the TAG.” 

Autosport felt that the team in red was ahead. “Ferrari look to be heading for their best season in years,” wrote Nigel Roebuck in Autosport’s season preview, just as the MP4/4 was finally hitting the track for the first time. And that’s the point. When the MP4/4 finally turned up at Imola the following week, Prost was immediately in the 1m28s and Senna did a 1m27.86s on the Wednesday afternoon. Senna would go on to take pole in Rio with a 1m28.096s lap, then topped Imola qualifying on 1m27.148s. Everything else, including the MP4/3B, was immediately uncompetitive. 

Jeffreys says that “it was never the plan” to race the MP4/3B, which would have been illegal anyway thanks to a rule change that demanded the driver’s feet be behind the front axle line. “When the MP4/4 turned up and blew everybody away it made the MP4/3B look like what it was – only as good as the best of the rest.” 

Nevertheless, the MP4/3B had more than done its job – and continued to do so. While MP4/4-03 became the test car in Europe after the Brazilian GP, the Japanese MP4/3B continued running until September, when MP4/4-03 arrived for the final 1988 tests.

“If we hadn’t had the MP4/3B it would have been a difficult learning process. The chassis was very different to the 4/4, but it was so much help on engine reliability, fuel settings and electronics settings” Neil Oatley

“It was a tremendous help,” argues Oatley. “It allowed us to develop the engine and the relationship with Ayrton. If we hadn’t had the MP4/3B it would have been a difficult learning process. The chassis was very different to the 4/4, but it was so much help on engine reliability, fuel settings and electronics settings.” 

Lall also believes it helped the game-changing MP4/4 hit the ground running despite a lack of mileage on the new car: “The MP4/3B allowed the MP4/4 to go straight into race distances because we knew the engine could do it. The mileage the 3B made available to the engineers for Honda and McLaren was a springboard for the 4/4.” 

The experience with the MP4/3B and the testing programme also showed the way forward for McLaren. “Having a second test team in Japan was a novelty and it continued,” adds Oatley. “It was invaluable for the V10 and V12 engines [for 1989-92] and other projects. It changed the whole philosophy of how we went testing.” 

Remarkably, the MP4/3 almost helped McLaren in yet another way. Plans were drawn up to use it to test Honda’s V10 for the switch to normally aspirated power in 1989. “We got a fair way through it and built a brand-new chassis, which is still here in the warehouse,” reveals Oatley of the MP4/3C. “But it became obvious how much quicker the 4/4 was, which is why we built two 4/4Bs…” 

When it finally arrived, the MP4/4 was miles ahead and kick-started a period of McLaren-Honda domination

Photo by: Motorsport Images

“You couldn’t go back,” asserts Lall. “We were the first team to have a bespoke test team. The way the 3B kicked it off was the stance and that became the level. Honda wanted more and more, and we wanted more and more.” 

Not only did the MP4/4 score more than three times the points of its nearest rival in 1988, but the McLaren-Honda partnership would rack up four championship doubles and 44 grand prix wins over five seasons. The MP4/3B didn’t score any of those, but deserves to be remembered for playing its part in one of F1’s most successful partnerships. 

The significance is not lost on Chris Vassilopoulos, whose family is the current custodian of MP4/3B-6. McLaren kept the car until 2019, when it came into John Stafford’s possession at Mouse Motors. McLaren Heritage partner Paul Lanzante Ltd was commissioned to get the car back to original, test-ready condition and the job was completed in 2022. 

Vassilopoulos didn’t go out looking for it but, with an interest in Marlboro-era McLarens and a collection that includes an MP4/4, he took the opportunity when McLaren started placing cars with collectors and marque enthusiasts. 

“All the cars were incredibly special for different reasons and there are so many changes in that period, with engines and technology, I think it was a golden era of GP racing,” he says. “Everyone gravitates to the MP4/4 because it was so dominant, but if you get into the detail you realise that didn’t happen by chance. The MP4/3B made all that possible.” 

Vassilopoulos, who is one day hoping to get behind the wheel, believes the MP4/3B should be seen. “The car feels like it could be used in great events,” he says. “There’s a groundswell for getting these cars used so people can see and enjoy them.” 

The last word goes to Pirro: “Test cars like this laid the foundations for something special.”  

How McLaren’s testing provided a new F1 lifeline

Pirro believes the experience gained at McLaren made him a better driver – and was the reason he raced in F1

Photo by: LAT Images via Getty Images

Emanuele Pirro had finished third in his second International F3000 campaign in 1986, but had given up his aspirations of getting to the top of the sport by the time McLaren came knocking. 

“It came when I’d put aside my F1 dreams completely,” explains the five-time Le Mans winner. “I could not get into F1 without bringing sponsorship, so I concentrated on touring cars in 1987 with BMW and I was a very happy man. 

“Towards the end of 1987 I got a call from my good friend Stefan Johansson and he said [McLaren sponsor] Philip Morris were doing a PR event in Arizona and needed a driver because Alain [Prost] couldn’t do it. I went along and drove the TAG car and then Ron Dennis said, ‘We are finalising a deal with Honda and we want to do a test programme in Japan, would you like to be the driver?’ 

“Japan seemed far away – more so than today – so it was quite a call. Ron said Honda would organise a drive in Japanese F3000, which wasn’t a route to F1 then but had some good, professional drivers, and I said I would do it.” 

“To have the privilege to be alongside special people – the commitment, attention to detail, motivating people, empowering the right people. This really helped me become better” Emanuele Pirro

Pirro would finish third in the 1988 Japanese F3000 standings and won the Suzuka round the following year before getting an F1 race seat at Benetton, replacing Johnny Herbert. Even during his time there, and while subsequently driving for Scuderia Italia, Pirro continued to test for McLaren.  

“Without this I would not have got to Formula 1,” he says of his McLaren work. “Sometimes opportunities are not where you think they are. I could have done something else instead of going to that Arizona event…” 

Pirro also believes the experience stood him in good stead for the rest of his career. “Driving for such a team with such drivers helped me to become a better racing driver,” adds the 64-year-old Italian.

“To have the privilege to be alongside special people – the commitment, attention to detail, motivating people, empowering the right people. Some things you can’t do but others you can and this really helped me become better. I set myself new standards, not only to describe the car but also to understand what is happening while you are driving. It was a wonderful four years for me.”

Autosport would like to thank McLaren, Mouse Motors and Chris Vassilopoulos

This article is one of many in the monthly Autosport magazine. For more premium content, take a look at the Summer 2026 issue and subscribe today.

Familiar and yet unusual: Honda lettering and sidepod snorkels reveal this is no ‘ordinary’ MP4/3

Photo by: Mouse Motors