Archive for December, 2006

Wanted – a new performance metric

Thursday, December 14th, 2006 Nigel Jones

In the bad old days, the two major performance concerns in CPU selection were whether a CPU had enough processing power and memory to get the job done. Although these are still issues, it’s a rare problem that requires more bandwidth and memory than can be provided by the CPU vendors.

By contrast, today, well over half of the systems I work on are battery powered, and so I find the major question I have when designing an embedded system is ‘how long will the battery last?’ If you can work this out from studying the data sheets of the various CPU vendors then you’re a better engineer than me.

Thus to solve this problem, I propose that we introduce a new performance metric – namely how much energy (Joules) does it take to perform a set of standard tasks. Rather than the usual bunch of quasi meaningful benchmarks, I’d like to see benchmarks such as:

  1. How much energy does it take to receive and transmit one thousand characters through an asynchronous serial port running at 38400 baud?
  2. How much energy does it take to perform a task switch using a standard RTOS such as uCOS-II?
  3. How much energy does it take to perform one thousand A2D conversions?
  4. How much energy does it take to execute a 64 tap FIR filter?

With metrics such as these, the task of choosing the best CPU (and compiler for that matter) would be made much easier. I’m quite prepared to let off the hook those vendors that aren’t selling CPUs aimed at the portable market. For the other guys (TI, Atmel, ARM etc) it’s time to step up to the wattmeter and be measured.


Wanted – .TEC password

Friday, December 8th, 2006 Nigel Jones

It’s time for my first rant – you have been warned!

I recently bought a new computer, complete with a gorgeous 24″ flat panel display. The flat panel supports a speaker bar – which I also bought. The installation instructions for the speaker bar are quite straightforward – align the tabs on the bar with the holes in the display, and push until the bar clicks in to place.

Well, on my system, there’s no click. The display seems to lack the spring loaded latch necessary for this to work.

I have now had four email exchanges with ‘technical support’. The first didn’t read what I wrote, the second told me that this was a big issue and would take several days to resolve, the third did a keyword search on ‘speaker bar’ and sent me a bunch of useless links, and the fourth decided that my problem was that I didn’t understand the installation instructions – and so sent me another copy of them.

In short, I’ve been treated like a moron.

I suspect that some / many / most people that contact technical support lack, ahem, technical acumen. Well, if you are reading this blog, the chances are you are not such a person. I also suspect that you’ve had a similar experience – which got me to thinking. What I need is a .TEC password. Just as Microsoft’s .NET password lets you manage your net identities, a .TEC password would tell the recipient that they are dealing with someone who really can, at the very least, align two tabs with their mating holes and push – and so should be treated accordingly.

Thanks for listening.



Monday, December 4th, 2006 Nigel Jones

As someone that has worked in telecomms, I was excited by the arrival of VOIP. However, after two years of variable quality, extended outages and just plain weird behaviour I’ve had it. It’s clear to me that VOIP just isn’t ready for prime time and so I have decided to pull the plug. The latest frustration – an inability to receive incoming calls for the last four days – with no resolution in sight. The technical support department informs me that it’s a ‘router programming error’. Whether they really mean a router configuration error, or a bug in the router firmware is unclear. Regardless, it’s presumably a tough enough problem that it can’t be fixed in four days.

The really bad news here is my experience when I tried to get Verizon to provide me with a POTS line. One of my prime reasons for jumping on VOIP as soon as I could was my feeling that Verizon was a dreadful company – one with questionable ethics and really awful customer service. Today, despite calling the number on the Verizon website for ‘add a new line’, I had to endure a voice prompted menu system and three different people before I could do the most mundane thing Verizon has to offer – order telephone service. For this privilege, Verizon is charging me a $44 start up fee (to plug a few numbers into a computer) and a cost double that offered by my VOIP provider. Apparently Verizon has not had its business suffer enough – yet.

So what’s the relevance of this tail of woe to embedded systems? Not much really, other than to note that when the latest and greatest doesn’t live up to its billing – one ends up with very annoyed customers. So next time marketing wants to over-hype what you can deliver, rein them in hard and fast. Your customers will thank you.