This thread discusses the Content article: Z-Wave versus Zigbee - A Quick Intro to the Z-Wave Protocol Followed By A Biased Editorial

I have worked in embedded software development for many years and have been doing some work on these two protocols a while ago.

The comment above is not really fair to Z-Wave.

Instead of finding the weakpoints of Z-Wave it's far more important to find the strengths.

One good advantage of Z-Wave is the fact it's driven by a centralized group managed by Zensys maintaining fairly strict compatibility between devices and with focus on SIMPLICITY.

Measured on quantity of features Z-Wave is not matching Zigbee but the market segment of residential home automation see little motivation in a bloated protocol and expensive components.

Too often protocols get bloated with needless features that might be something you can optionally choose.

What happens when things are not mandatory and the group defining protocol requirements just include all member requests as options?

The answer is usually that compatibility sucks and QA departments scream when they realize the complexity of testing various generations of products from various vendors with various options chosen. Protocols with many optional features are long-term nightmares in maintaining compatibility.

Z-Wave is popular because it's simple and is doing exactly what it needs to do. Adding more features to Z-Wave will only insignificantly impact the usability of it as there are only so many ways you can control your light, aircondition, home security etc.

So the point is...for Zigbee to succeed against Z-Wave it must have a very clean and simple mandatory base protocol profile that nobody violates.

MINIMIZING options should be a primary goal for residential use because options lead to compatibility nightmares over time.

As I often say in my profession...real men are the ones who can keep things simple by firmly saying "no" to all non-essential requirement requests.
Zensys seem to have a better handle of this I believe. Am I wrong in that?

Thanks for the comment on the post. You brought up many good points and I do think that Zigbee does suffer from complexity due to feature creep. Also Z-Wave does do a good job of keeping things simple for small networks, such as using source routing, limiting the maximum depth of the network and the number of router hops.

On the bad side, the Z-Wave spec is completely controlled by Zensys where they provide both the chips and the software. A closed environment prevents the diversity and competition that would occur in a more open environment where an industry standard radio and networking protocol could be used. On the hardware side, the 802.15.4 radio prices should fall faster than the Zensys radios. And the diversity of the 2nd and 3rd generation radios that are coming out now are much more sophisticated than the radios that were even available when I wrote that post.

On the protocol side, the Zigbee spec is big and fairly bloated. I think everyone pretty much agrees on that. From a compliance point of view, the Alliance has tightened up the certification testing and the profile compliance for the 2007 specs.

QUOTE:
"The ZigBee specification has a number of options, which, if exercised in different ways by different vendors, will hamper both compliance testing activities and future product interoperability. This document, which is, for the most part, a set of restrictions on the Protocol Implementation Conformance Statement (PICS) documents corresponding to the three main sub-clauses of the specification, further restricts those options so as to promote interoperability and testability."

That quote is from the Zigbee Feature Set Profile documents for both Zigbee 2007 and the Zigbee 2007 Pro. So the Alliance is trying to address and improve the issues of maintaining interoperability.

As for the feature set, it may be overkill for the Home Automation Profile as compared to Z-Wave and I do agree that Z-Wave does do home automation well. In fact, I think Z-Wave will have less competition from Zigbee in Home Automation than it will from the RF4CE group that has recently been formed. This group has the benefit of learning from the good and bad points of both Z-Wave and Zigbee and enjoys backing from some strong manufacturers. They also will be using 802.15.4 radios which will enjoy the same price curves as Zigbee, 6LoWPAN, ISA100, and Wireless Hart.

On the plus side, Zigbee is moving ahead in creating interoperability profiles for a lot of applications, with the "Security and Safety", "Wireless Sensor Network", and other profiles coming up soon. So in the sense that the spec is bloated, I guess it depends on the application domain. If they were just targeting home automation or smart energy, then they could definitely lean up on a lot of things. But since the application domain is quite wide, then they are pretty much providing a framework of services and each profile will consume different parts of those services.

Also, the spec is constantly pruning redundant or unused areas as can be seen from the original 2004 spec to the 2006 and now the 2007 versions. It would be nice if they informed the stack implementors about the changes too, though.

I liked the editorial very much, thanks! I'm probably very biaised too since I'm involved in the ZigBee Alliance but I found it quite balanced.

The comment below regarding the simplicity of Z-Wave is quite fair and this is probably one of the downsides of ZigBee, at least the consequence of a willingness to address many different applications using the same protocol. Relying on a consortium of 250+ members is obviously much more difficult to manage and direct than driving alone a spec. That said, the main issues about ZigBee and Z-Wave are more strategic than technical. My company is a large OEM producing millions of devices that need sometimes to operate for 10 to 20 years (not consumer electronics). We need a global, worldwide standard, and we need multisourcing. Competition is key to boost innovation and decrease cost. That's the real nice thing about ZigBee.

We have had experience with both ZigBee and Z-Wave. Although Z-Wave has a well designed stack, hardware is the real weak point. Zensys cannot be competitive (in terms of performance and cost) compared to the 802.15.4 players. Whether ZigBee will dominate or not the WSN space is still a question mark, but what is now sure is that 802.15.4 is the real reference driving the most promising standards (ZigBee, ISA100, WiHART, 6loWPAN).

That was my one cent contribution. Congratulations for this site and good luck with the stack!

I think that the price of the zensys ic is very expensive too.

The open standards create new oportunities and the world is adopting.