Is Yamaha Using A Seamless Gearbox? The Data Says No

Ask Jorge Lorenzo if there is one thing which the Yamaha needs to allow him to compete with the Hondas, and he will tell you it is a seamless gearbox. The system used by Honda on the RC213V allows the riders to shift gear while the bike is still leaned over, without upsetting the machine. It is an important factor in the Honda's better drive out of corners, as Dani Pedrosa, Marc Marquez, Stefan Bradl and Alvaro Bautista can shift gear earlier and make optimum use of the rev range to accelerate harder.

That Yamaha is working on a seamless gearbox is no secret, with Yamaha's test riders currently racking up the kilometers around tracks in Japan, testing the reliability of the maintenance-intensive system to the limit before using it in a race. Recently, however, Spanish magazine SoloMoto published an article suggesting that Yamaha has already been using its new seamless gearbox since the beginning of the season. In evidence, the magazine pointed to an apparent difference in fuel consumption between the factory Yamahas and the satellite bike of Cal Crutchlow. While both Cal Crutchlow and Valentino Rossi made mistakes at Qatar, only Rossi was able to recover, and then battle with Marc Marquez for the podium. The theory put forward by SoloMoto was that the smoother transition between gears gave both better drive and lower fuel consumption, as the ignition is cut for a much shorter period, wasting less of the limited gasoline the MotoGP bikes are allowed.

My own enquiries to check whether Yamaha was using a seamless gearbox or not always received the same answer: no, Yamaha is not using the seamless gearbox. The reason given was simple: with Jorge Lorenzo defending his title and Valentino Rossi in the race for the championship, they simply cannot afford to have a single DNF down to a mechanical failure of the new-fangled seamless gearbox. The risks involved were just too great, especially when taking the reduction in engine allowance into account, with just five engines allowed all season, down from six in 2012.

To test this denial, I went out to the side of the track on Friday morning at Jerez to record the bikes as they went by. I sat at the exit of Turn 10, Peluqui, and recorded the bikes as they accelerated towards Turn 11. It is a spot where they change gear once, before braking briefly for Turn 11 and then powering on to Turn 12 and the final short straight and hairpin.

Once I had enough recordings, I analyzed the sound clips in Audacity, an open source audio software package, measuring the length of time the gear changes last. Given enough samples, it is a relatively simple task, as the point at which the ignition is cut is clear from the audio (see screenshots below). I put those timings into a spreadsheet, and then averaged them, both for each rider, and for the Yamaha and the Honda as well. Below are the times I registered for Valentino Rossi and Jorge Lorenzo on the factory Yamahas, and Stefan Bradl and Marc Marquez on the Hondas. Both the LCR Honda of Bradl and Marquez' Repsol Honda use Honda's seamless transition.

Length of gear change, in seconds:

  Rossi Lorenzo Bradl Marquez
  0.043 0.038 0.008 0.012
  0.042 0.032 0.008 0.006
  0.038 0.031 0.007 0.014
  0.043     0.007
Average 0.042 0.034 0.008 0.010

Averages for the Honda and the Yamaha:

Yamaha Honda Difference
0.038 0.009 -0.029

What is clear from the timings is that the Yamaha is still significantly slower in changing gears than the Honda. The Yamaha takes an average of around 38 milliseconds, while the Honda takes just 9 milliseconds, a difference of nearly three hundredths of a second. These results are nearly identical to the results I found when I first checked the data at Qatar back in 2011, when it became clear that Honda was using the seamless gearbox.

So what conclusion can we draw from this data? It seems clear to me that Yamaha are not using a seamless gearbox just yet, and that Wilco Zeelenberg was telling the truth when he denied to me that Yamaha were using it. The data from the sound clips is almost irrefutable, as the length of time each gear change is taking is almost the same as it was two years' ago. If I am wrong, and Yamaha are using a seamless gearbox, then it is not a particularly good one.

What is clear from both the sound recordings and standing trackside is that the electronics package has been modified to handle gear changes better. Where once  each gear change was accompanied by an enormous bang, as fuel hitting the hot exhaust exploded, that noise has been greatly reduced. Less fuel is being wasted in those gear changes, for certain. But a new electronics package does not a seamless gearbox make. The Yamaha riders will have to wait.



Back to top


So on average, about 34 gear changes amount to one second. I'm not knowledgeable enough to take the math further and there are certainly many other variables, but it looks to me that Honda could be in very serious trouble whenever Yamaha does get their box sorted out.

Good point. And I think what is more important than the lap time that could be "gained" from a seamless gearbox would be the ability to pass the Hondas on conrner exits and straights and not just on the brakes as is the case 90% of the time at present. Basically nulifying the Honda advantage that we saw cause problems for Crutchlow at Qatar.

In terms of real time the seamless shift saves over a lap the difference is very very small. IMO the benefits are more relevant during the race, where the seamless shift just makes it easier to ride at the limit. A track instructor I had once said if you break up the total concentration a rider has to use into 10 tickets, where he spends 2 tickets on his line and reference points, 3 on body position/bike control, 3 on throttle, brakes, shift and a couple on race raft, then if the seamless shift means he can spend half a ticket less concentrating on shifting that means he can use that elsewhere which can help when you are riding on the very limit for 40-45 minutes. To me that's an area Yamaha riders have had an edge over the rest even without the seamless shift with the M1s sweet engine, electronics and chassis making life easier for the Yam riders during the race, equating to less mistakes, higher consistency and possibly a little less fatigue for the end of the race. Where over a single lap on soft rubber the Honda may have a slight edge at some tracks, over race distance the Yam is superior, as it's domination of the last decades titles will attest. As others have said, give Jorge a seamless shift as well and the Yamaha may well be unbeatable!

"Where over a single lap on soft rubber the Honda may have a slight edge at some tracks, over race distance the Yam is superior, as it's domination of the last decades titles will attest."

I suggest that you should check your statistics.

Since Yamaha introduced the long stroke inline 4 engine they've won 6 of the last 8 titles, with Ducati and Honda taking just one each. I call that domination.

You are wrong! You wrote: "over race distance the Yam is superior, as it's domination of the last decades titles will attest.", and what you wrote is simply not true, "decades" - what decades? - check you're stats again! There is no decade in WGP racing in which Yamaha is better than Honda! I don't count '70 because Honda didn't compete in that decade.

Then you wrote: "Since Yamaha introduced the long stroke inline 4 engine they've won 6 of the last 8 titles, with Ducati and Honda taking just one each. I call that domination." Wrong again, check you're stats!!! You have surely meant on rider titles and not manufacturer titles, well, as you know - rider title goes to rider and manufacturer title goes to manufacturer. They shouldn't be confused. For example, Valentino Rossi has won rider title in 2004, but Honda has won manufacturer title in that year.

Here is the stats from 2000 to 2011:

Season Category Constructor
2011 MotoGP Honda
2010 MotoGP Yamaha
2009 MotoGP Yamaha
2008 MotoGP Yamaha
2007 MotoGP Ducati
2006 MotoGP Honda
2005 MotoGP Yamaha
2004 MotoGP Honda
2003 MotoGP Honda
2002 MotoGP Honda
2001 500cc Honda
2000 500cc Yamaha

For more info please check this link below!

But I did forget Hayden's title in 06. That makes it 6 of the past 9 titles for Yamaha since they introduced the long stroke engine, with Honda nabbing 2 and Ducati just 1.. Over the full decade since 2003 It's Yamaha 6, Honda 3, Ducati 1, so Yamaha still have twice as many titles as Honda. Whether that fits your definition of Domination or not thats a very convincing lead to Yamaha.

P.S. Suzuki won the 2000 title too. Might want to check your own link there bro!

No, you're not right, the Yamaha didn't won the titles in 2004 and in 2012, it was Honda, in both year. Over the full decade starting from 2003 to 2012, the manufacturer (or constructor) champioship score is; Honda 5, Yamaha 4, and Ducati 1.

Here is the stats from 2003 to 2012:

Season Category Constructor
2012 MotoGP Honda
2011 MotoGP Honda
2010 MotoGP Yamaha
2009 MotoGP Yamaha
2008 MotoGP Yamaha
2007 MotoGP Ducati
2006 MotoGP Honda
2005 MotoGP Yamaha
2004 MotoGP Honda
2003 MotoGP Honda

You don't seem to understand. No matter, I will write again. Look, the rider title goes only to rider as his personal achievement - not as manufacturers, and munufacturer (or constructor) title goes only to manufacturer as his. It seems that you're mixing this two things.

Here is the clarification:

The Riders' World Championship is awarded to the most successful rider over a season, as determined by a points system based on Grand Prix results.

The Constructors' World Championship is awarded to the most successful constructor over a season, as determined by a points system based on Grand Prix results. Only the highest-scoring rider in each race for each constructor contributing points towards the Championship. The winner of the constructor's world championship is not necessarily the bike used by the riders' world champion. For example: In 2004, Valentino Rossi who rode a Yamaha bike won the riders' world championship, but in the constructors' standings, Honda have higher points than Yamaha, therefore Honda won constructor's world championship.

P.S. My link is good but you do not seem to know how to use it. Suzuki didn't won the title 2000. It was Kenny Roberts Jr. and the winning constructor was Yamaha, bro!

"While Crutchlow ran out of fuel on the cool down lap, Valentino Rossi was able to recover from a mistake, and then battle with Marc Marquez for the podium."

1. Wasnt it Rossi who ran out of fuel on the cool down lap and hitched a ride from Iannone?

2. Whether the ignition/spark is killed for .009 or .043 seconds, with the thottle wide open, does that really improve mileage? The shift make be quicker, but isnt the same amout of fuel being delivered for the same amount of time? Meaning, less fuel may be "wasted" but just as much is being consumed.

Two good points.

1. You are correct, I got my memories backwards. I'll fix it in the text.

2. When the ignition is killed, fuel is being injected without creating drive. The bike stops accelerating, and more fuel is needed to get it back up to speed. When the time the ignition is killed is shortened, the bike slows less, and so has less speed to try to make up again. So the same amount of fuel is consumed, but the resulting speed is higher.

Fuel will also be cut during the ignition cut time. The engine would take a few cycles to recover, this is what Marquez's team did last year on a control electronics package. Makes sense a factory team would do it with blue sky electronics

Whether ignition, fuel, or both are cut, what suffers most is momentum. That momentum has to be regained, and the shorter the cut, the less momentum is lost in the first place. This, I believe, is where the real difference lies, and where fuel is saved.

Great stuff using available tech to suss out differences.

So, how fast do the Ducatis, ART, and other bikes shift?

If someone recorded bikes going past (& labeled the recording), the analysis could be done by anyone that could download the recordings.

Terrific analysis, I think similar things have been done in the past quite a bit, not only MotoGP but F1 to analyse engines and gearboxes. I really enjoy the scientific approach and the work done to assemble this piece, thank you for an excellent article.

Honda no doubt already know this kind of Yamaha data too, so they must be already looking for the Next Big Thing in the evolution of the RCV... or maybe they have already found it - Honda Factory Part # MM93

Honda isn't a baby lost in the woods.

Evan Williams wrote:

"It’s generally accepted that when things are rolling at HRC, they are a couple of steps ahead in their development. When it becomes obvious the riders need more on the track, they don’t cook something new up. It’s already finished and sitting on a shelf, waiting to be called into duty. The engineers are done with it and working on something else."

It's not really momentum, it's the time taken for the engine to restore full power after cutting the ignition. If you cut fuel as well there is no recovery period.If you don't cut the throttle you don't loose momentum.
What is amazing is HRC has achieved faster shifting than electromechanical shifting in F1

I would not like to say that my timings are perfectly precise, I am having to make educated guesses at exactly where the shift stops and starts, so I could be a few milliseconds out either way. I'm very confident in the magnitude of the difference, but the precise length of the shift has a small error margin.

David you can also analyze the frequency of the audio signal and it should give a clue as to what bikes rev to what RPMs as well. As a musician with a good ear, I can tell ,but it would be very interesting to analyte the rev range of the bikes. The diference in pitch generated by a screamer engine would be higer of course but doing simple math you can also tell what the RPMs of a banger engine si, as it seems the frec would be half of that of a screamer at the same revs.....or so I think :p

Unfortunately, the terms of my media pass forbid me from uploading sound recordings made at the track. Sorry...

When on a high throttle opening and/or in a 'medium' speed, or faster corner wind and rolling resistance/friction (these bikes are often not in line in corners and tyres may be spinning) will quickly slow the bike down if power is reduced. This is probably imperceptible to anyone but the rider and the engineer with the data, but when we are talking thousandths of a second it all adds up.
The consistency of these riders is phenomenal and someone like Lorenzo will unquestionably capitalise on that - both directly by using the time/power available to accelerate more quickly for longer (it builds with every change) and also by finding new ways of riding the bike to make use of this new ability. The last point is pure guesswork but the former is not.
I hate to think how much many is being spent on this technology because it all seems pretty pointless - like carbon brakes it has no meaning beyond the track.
If DCG was allowed, or seamless banned (just specify a minimum gear-change ignition interruption as noted by David if they don't know what the technology is) then they could spend a small part of the saving on having one male umbrella person for each female one....