What’s faster – A QNX CAR or a Genivi?

August 24, 2009

I’m a little overdue in writing this but at a recent tradeshow we booked a meeting room. We had a parade of automotive Tier one suppliers, OEMs and partners come through the meeting room to see the latest on the QNX CAR reference platform.

What was different this time is while we were showing the latest integrations (Pandora, Chumby etc), we were also demoing the first Tier one implementation based on QNX CAR. The Tier one, who would prefer to remain nameless, allowed us to show off their fully featured mid-level infotainment unit prototype. They took QNX CAR, added their preferred navigation and speech recognition technology and customized the HMI. It was a very slick unit.

What’s the big deal you might ask. Tier ones develop prototype systems all the time?

This Tier one went from a standing start to a fully functional prototype in under four months. That’s what got everyone excited at the show. That’s what you can do with QNX CAR.

I wonder how long it would take if you tried to do this with Genivi?

Romain


Genivi and automotive platforms

March 24, 2009

 

At Cebit, QNX, Microsoft and Genivi all revealed their visions for a common automotive reference platform that would cut costs, accelerate time to market, reduce costs and just make the world a better place. I won’t take the time to review each offering in detail as there has already been a fair bit of coverage from the media.   

 

 

In these articles, one thing struck me as particularly interesting. People are now identifying software as the single most important element in automotive systems going forward. I agree wholeheartedly. I spend a lot of my time lining up third party software vendors to meet the myriad requirements demanded by OEMs for upcoming model years.  It is no trivial task. Bluetooth connectivity, navigation, points of interest, remote update, internet radio, multi-media, device connectivity, speech recognition – the list goes on and on. Thankfully there are software companies that specialize in each of these areas.

 

The notion of a common automotive reference platform comes up a lot these days, even more so since the economy tanked and everyone has to do more with less. Ideally the vendors serving each of these very different areas would work cooperatively to shoulder the burden of integration and testing – a consortium of like minded industry players coming together to build something that could be used by all. A truly open, standards-based organization where the ultimate output would allow the automotive industry to choose exactly what functionality, features and vendors it wanted to work with. That’s very powerful stuff.

 

Genivi is a consortium that is being driven by a handful of automotive Tier ones and a couple of OEMs but only one silicon vendor and one software vendor. There is only one software vendor in the consortium today. My understanding of automotive requirements suggests that even a huge software giant couldn’t possibly hope to address everything needed, even if they had the next 50 years to get it done. I guess that’s where this community they talk about will have to help a lot.

 

I’m not suggesting that QNX CAR is the perfect solution either but at least it does encompass a rapidly growing number of software vendors and offers support for all the major automotive silicon choices out there today.  It is not a standard, per se, although the underlying operating system is POSIX compliant and brings the benefit of providing a standard API. It is open to pretty much everyone and its goal is to provide a set of pre-integrated, auto hardened technologies under a business model that promotes its use for prototyping and product development.

 

Now to be fair, I don’t actually know a lot about the details of how Genivi plans to roll out their platform. I’m not invited.

 

Romain