Meindert Sprang wrote:

"Yme Bosma" wrote in message
om...

Hi there,

It's been discussed in this forum before, but over at Ocean Navigator
they have published an article that states the NMEA 2000 standard is
'picking up steam'. Subscription is required, so here some quotes from
the article:

http://www.panbo.com/yae/archives/we...22.html#000356

I'm curious to know what you all think of this standard and wether
it's really gaining any traction. Or do you already see any rival
(non-proprietary) 'standards'/alternatives emerging?


Based on CAN, it is a robust standard, but requires quite some protocol
overhead since CAN can only transmit small datapackets.

My biggest problem is the cost involved. To get your first product on the
market, you have to buy the standard documents and test suites and apply for
a vendor- and product ID. This will cost $10,500 total, quite a hurdle for
small manufacturers.

My 'all time favourite' would be a marriage between NMEA-0813 and SeaTalk
and some other features. Use a CAN bus driver, (passive '1' level/active '0'
level), use the same text-based NMEA type sentences (easy to debug), use the
collision detect feature of SeaTalk and increase the speed to a few hundred
kbit/second. And when need be, switch to a binary variant of NMEA.

Meindert



I can't agree. Hardware is cheap to develop relative to software. A
handful of hardware engineers can keep armies of software developers
busy indefinately. This is epecially true for low run items like boat
and ship electronics.

Therefore, my favorite solution is NMEA 0183 sentences, text readable
ASCII and all, over IP multicast Class E addresses with some entity
keeping track of an IP address registry. At 100BaseTX rates each device
can put out whatever sentences it wants at whatever rate it wants and
the consuming devices can simply subscribe to those multicast IPs it is
interested in. Streaming video and all. The silicon is there; the
software is there, too. Someone else is designing the tablet PCs. So
what if it isn't totally waterproof to that 10 minute fire hose blast
standard. With COTS HW/SW the cost will be 1/3 or less of the purpose
built equipment. Buy 2; stow one in a waterproof bag with dessicant,
and pocket the differential.

This has got to be the fastest way to market with a the lowest
development cost to be spread over the installed base.

I worked in the development arm of a world reknowned R&D lab for nearly
30 years. Been there; done that.

Joe Wood