MotoGP Tech Director Corrado Cecchinelli: "The Unified Software Is At About 10% Of Its Potential"

The Valencia test saw two of the major changes for the 2016 MotoGP season make their public debut. Testing with the Michelin tires has been going on since Sepang in February, though Valencia was the first time the riders were allowed to speak about them. The 2016 spec electronics package, the so-called unified software, made its first appearance on track in public, it having only previously been used by test riders at private tests.

The unified software was met with a mixed reception among the riders who used it. Though everyone said it was very much a step back in time, some riders were fairly happy with it, while others were much less so. The main criticisms leveled at it were that it reacted inconsistently, and it was difficult to get a handle on exactly how the software would react from one lap to the next. Whether this is fundamental to the software, or related to the fact that this was the first real outing on the software, and the factory software and engineers needed time to sort the electronics out will only become clear as the season progresses.

To find out more about the software, we spoke to Corrado Cecchinelli, MotoGP's Director of Technology, who has overseen the development of the unified software, after the first day of testing with the electronics. The software was still very much under development, Cecchinelli said. "This is a development version, but you have to understand that work will never be over. Our intention is to improve and follow the manufacturers request but of course the pace at which we do that is higher now. For instance we are using a version here that has already been superceded by a version that is ready for the manufacturers in Sepang."

The inconsistency reported by riders was largely down to the fact that the teams had so little experience with the software, Cecchinelli said. "I would expect that the tuning at this moment is about ten percent of the potential. Everybody is very far from using the best potential of this software." Despite this, he said, the goal was not to replicate the performance of the software the factories had been using until now. "We all have to understand and share that we are in the project of making the best unified software possible, which by definition, will not be as good as the best factory software possible. If you want it’s a bigger and more challenging project, but the way the bike will behave will be a bit poorer of course to what the factory software could do. This gap to me when everything is ok and we are using 100 percent the potential with one of the future releases, this gap, compared to a theoretical ideal factory software is hundredths or tenths of seconds a lap. Here [at Valencia] we are in tenths to seconds per lap."

Cecchinelli had some sympathy for the complaints of the riders. "All the riders comments makes sense to me. From a rider perspective I believe it’s a step back. Today, for sure it's a step back, but they have to understand that it will be the same tomorrow as well. It will work better but they will be used to it. The potential performance will still be a bit less." 

Though Cecchinelli was open to feedback from the riders, the most important link for him, and for Magneti Marelli, who develop the software, was with the engineers and crew chiefs from the teams. They were able to convert rider feedback into a format more useful for the software engineers. "I think that the feedback which is most important for us is the feedback from the manufacurers engineers. They have the data and riders’ opinion and they can put them together and turn them into something usable from an engineering side. Of course the rider can only say to me this is a step back. That is not so useful to make it better. The engineer must translate it into something useful for us. I mean maybe it could be the wheelie, things like this we need to understand better. It makes sense that the riders feel it worse than two days ago but it’s not enough to make it better."  

Cecchinelli emphasized that the first avenue of attack for the inconsistency reported by the riders would be through optimizing the existing software, and using it better. "I know you can understand that this [inconsistency] can’t be something related to software. This is a matter of tuning. There is no reason why it should be inconsistent from corner to corner. It’s just a matter of tuning it to the best in corner to corner." The most surprising aspect to him was the complaints from riders about inconsistency in the same corner. "I think we have to understand it better because there is no reason for it. There’s no reason in general for software to be inconsistent. We need to understand what they mean because the language is different. The language of bits is different to the language of riders, which is sometimes rude! This is why I’m not that worried. So far I didn’t hear of anything that was caused by really wrong software."  

A new version of the software is due to be used at the Sepang tests, but the main objective of updates at the moment is to reduce the chance of error and make the software safer. "The version for Sepang is actually better for failsafe aspects. There are more recovery strategies and things like that." The new version of the software would not improve performance, but that does not mean that the bikes will not all be going faster at the test, Cecchinelli said. "I don’t expect the performance to be better by itself. It must be better because everybody is more used to it and the calibration is better. There is no reason to go faster because of the software. For me there is a strong reason to go faster because everyone is more used to it. At the moment we are so far in the tuning process that you can find out things here so big that work in Sepang as well." 

Cecchinelli also gave some insight into the planning for the unified software. "The latest release schedule was November sixth and that was for Sepang and it is done. Now we’re making a meeting in December tenth and then we will make a decision about the next release, which will be after race one I guess, depending on what we’re putting into it. I cannot say now because I don’t really know what we’ll decide to put into that release. To me the idea is that the version from Sepang will probably be the one for race one. That is my guess now. You understand it’s sort of a risky guess because we are in a very quick process but this is my guess now."

The biggest question mark Cecchinelli was left with was how the software would work with the Michelin tires, which will be used for 2016. Some factories and teams were still using the 2015 software, which he felt was a wast of time. The new tires would make a big difference, he explained. "It brings a lot of difference in the calibration process because of course the software contains in some way a tire model. I mean every strategy refers to the tire behavior, after all, so calibration changes a lot but not the software strategies. To me it’s wrong to work here with the factory software with the Michelin tires because all the work you are doing. There is really no point. I’m not doing that job but it’s just to say there is a lot of calibration job related to the tire. Just imagine the profile. If you consider the software and the strategy there is no reason to change it. There is no reason to change it depending on the tire. The numbers you put into the strategies change a lot."

 

 

Tweet Button: 

Back to top

Comments

Rather than 'improve' the software I would prefer to see the teams left to dial back the ECU-controlled areas that are causing the inconsistencies, and transfer the responsibility for that to the riders' wrists.

David, any news on what's happening with the factories using their own inertial packages to fool/dodge around the control ECU? Your article a while back on the prospect of 'piggyback ECUs' was interesting. I think it's relevant here because the loss or degradation of the ability to judge the bike's position on the track, or angle relative to track camber etc, might feed into these inconsistencies in the behaviour, in a given corner, from one lap to the next that riders are complaining about.

they wil need a half year to get the electronics to get 100% and they wil find some ways to get that.
dont forget they have also relearn again on the michelin tyres ..

cant wait for 2016 we will see soon enough wo has the best package on bike and electronics

I'm not so sure I agree with Cecchinelli on working with the Factory Software and Michelin Tyres. I think it was Suzuki which stuck with their software and tried the new tyres. Going to First Principles, I'd always aim to change the minimal amount of variables. Trying the Tyres with the Factory Software lets them understand the difference the tyre makes against a known quantity - i.e. the bike + current electronics package. Changing the Electronics + Tyres would potentially leave people blaming one thing or another, that data Suzuki has got may prove useful over the winter for analysis.

I think it's a complex task for all the teams, but I think one point David has continually raised remains true and solid. The Factories ultimately will be up the front because they have more resources. Changing perhaps the two most major things you could, at the same time will be tricky to manage from an engineering perspective. The paddock has some very bright people in it, but never the less, some will do better than others, and the teams who can afford the most thinking time are likely to be nearer the sharp end. With change comes disruption though, so it's not impossible for smaller teams to work in a way that others haven't quite thought about and capitalise on it. I'm looking forward to 2016 already, and that's before Moto3/Moto2 which also look like they'll be excellent.

I thought exactly the same thing. To me it seems a logical move not to change these two things at once, because it would make evaluation of the tyres very complex and confusing. And the new tyres (not just from a very different manufacturer, but also different sizes!) make such a profound difference for suspension set-up and chassis geometry, that you really want to get a clear picture of that. When you know the difference the tyres make, it will be much clearer to see the difference the unified software makes, and what needs to be done with it.

But that may just be the reason for Corrado Cecchinelli that he wants the teams to change both things at the same time. That way they won't be able to isolate just how much of a step back the unified software is. Either that, or he does not have much of an insight in how to analyse and develop stuff. Which seems unlikely.

To be honest, I am not that impressed by his interviews. Maybe it's just his way of speaking English, but also in earlier interviews I found his explanations vague and inconsistent. It sounds more like a politician trying to sell a bad new plan to the public by using way too many words and vague logic.
It could also be that I am just too stupid, of course.

Yes, since Dorna wants everyone, or more accurately the factories, going slower, then changing a bunch of things at the same time will definitely help their effort.

I wonder how much of the vagueness is due to English being his second language. Statements like "Today, for sure it's a step back, but they have to understand that it will be the same tomorrow as well. It will work better but they will be used to it." don't really inspire confidence but can be due to translation.

Unfortunately, statments like this: '"I know you can understand that this [inconsistency] can’t be something related to software. This is a matter of tuning. There is no reason why it should be inconsistent from corner to corner. It’s just a matter of tuning it to the best in corner to corner.' can't be blamed on translation issues but can stem from a lack of knowledge of complex software systems. What can seem to be inconsistencies can actually be software bugs. To not even include the possibility of a bug makes me think he has a poor understanding of the software he is responsible for overseeing. This opinion has been reinforced by every interview I read about the topic.

Chris

This is just the same as all the other rules in motogp; fuel limits, tyre restrictions, engine restrictions, etc. A truly unrestricted prototype series would be interesting, but two things would come:

1) The team with the most money would always win. This happens a fair bit already, but there's no way a smaller team could compete if Honda were allowed to build any bike they wanted.

2) Safety would be compromised.

1. Actually, having more rules benefits those with more resources to spend on the diminishing returns. Having less rules would have the paradoxical effect of greatly reducing the asymptote on the rate of of diminishing returns. That is, less rules makes racing cheaper.

2. Maybe safety rules are the only rules the sanctioning bodies should be concerned about. Not technical regulations.

... by definition is supposed to stop "progress," lower costs long term and help level the playing field across the board.

I'm not saying it will do that, but that's the general idea.

if you watched all those races back then with Roberts, etc. You realize that they were putting on a show... It was entertainment. I think I remember seeing Roberts et al lap American Vintage Racer Ace, Dave Roper, who was on a bevel drive Ducati. They knew the field was thin and decided to have fun for most of the races, and then raced hard for the last lap or two...

What a ridiculous statement. Two months ahead of the season and the man proudly states that "The Unified Software Is At About 10% Of Its Potential"!
Is this supposed to instill hope for a thrilling 2016 or does it impute something untoward for the man and his planning abilities? A lot of what happen on the tracks next year will be blamed on this (or similar) rather ungainly percentage of untapped, underutilized potential.
Not blaming the idea, mind you.

I don't mind at all the fuzz, it's actually good to rattle the tree a little bit here & there, and see how does the riders & teams adapt, I bet the aliens will still be the top 4 but the order of the aliens may vary.

Could be a great season if it turns out the "humans" can get a little closer to the 4 aliens.

10% leaves a lot of headroom for development. With, effectively, new bikes on new tyres, and the probability that this software has not really been fully addressed by the factories, the development potential is significant. I suspect that's what he means, not that its 10% complete. If the reverse was true, and it was already at 90%, I think everyone would be worried at the lack of potential.