Comments on: Electric Imp aims to make the Internet of Things devilishly simple

By: Mark

Mark — Tue, 19 Mar 2013 13:31:35 +0000

I use the imp and like it a lot. I think the fit for this product is more likely smaller companies or others who strategically decide to focus on other areas. I work for a really large company and use the imp for test machinery rather than end product. I also use it for hobby.

the imp has really good RF performance, maintaining connections easily even though I have placed it near solenoid valves, underneath another circuit board containing several relays, and behind my machine, under wire shelves in the back of a shop.

It is really easy for writing and testing code. I can throw code at it from my office as it run at home and if I write a bug, the IDE gives me the line# of the function and the calling functions, the imp reboots and I can again send code.

I like being able to move the imp card between projects; this partially address the issue of cost. For example, I plan to make a connection to my treadmill. It is only used once per day so the imp card can be present for the workout to track certain metrics and then deployed elsewhere.

Mark

By: Hugo Fiennes

Hugo Fiennes — Fri, 15 Mar 2013 16:40:22 +0000

There are many misconceptions about what makes “low power”. Yes, WiFi isn’t as low power as some simpler wireless standards, but the imp does enable truly low power operation in many scenarios – eg, we can run an environmental sensor checking in every 15 minutes for 3+ years on 2xAA cells – or indeed powered from a small solar cell and a supercap. Would a zigbee sensor do better? Very possibly, but it would need a gateway, and that’s infrastructure many people don’t want to deal with. With an imp, any WiFi enabled home or business can install such a sensor in a couple of minutes.

The imp solution doesn’t fit every use case – but then again, neither do other solutions. The manufacturers we’re working with are very happy with the end-to-end solution we provide and what it allows them to deliver in their products.

By: smarthomeman

smarthomeman — Fri, 15 Mar 2013 15:30:46 +0000

I agree with Fred. This sounds mad. It’s not really an accurate understanding of the problems of IoT, and it’s certainly a seriously bad solution for most applications. Apart from anything else, WiFi is too power-hungry. Most of these devices will only need a few bits/second of bandwidth, and can use much lower power devices. With dozens or hundreds of devices in a typical future home, WiFi is a total no-no; it will destroy the planet.

OpenDCU.org takes a better approach, I think.

Having said that, there is a small class of high-bandwidth applications for which this might be a good solution. I hope the Imp guys will join the OpenDCU initiative and make their stuff compatible.

By: Fred Vargas

Fred Vargas — Thu, 14 Mar 2013 15:53:56 +0000

There are **much** more affordable, low power and flexible ways to get wifi on your pcb. This is a classic case of a product in search of a problem. The team might make a great fit a big company with a clear focus for them but independently, on their own they couldn’t build a succesful product to save their lives.