Saturday, November 13, 2010

mbed - Big Boys Arduino!

Hack-a-day described mbed as "Arduino. On steroids. With claws and fangs.", what's not to like about that?

The mbed is a small module with 0.1" pins for mounting on a breadboard. It has an NXP (Philips) LPC1768 Cortex M3 ARM 100 MHz processor which has 512KB Flash, 64KB RAM and a stack of goodies including Ethernet, USB OTG, SPI, I2C, UART, CAN, PWM, ADC, DAC and GPIO (ordinary General Purpose I/O pins).

The development tools are unusual in that you don't install anything, when you register your shiny new mbed, you get an account with your own workspace and access to the web-based compiler and libraries.

To register, you just hook the mbed up to a USB port on your PC (cable provided) and click on the MBED.HTM file in the flash-drive that appears (being web-based it works for Windows, Linux and Mac of course). Less than a minute later (if you have a user name thought up) you can be set up and ready to go.

To run the demo 'Blinking LED' program just click on the download button and save it to the flash-drive your mbed appears to be. Press the reset button on the mbed and 'Presto!' the latest file to be downloaded is executed, it really is that easy.

The demonstration video to show the power and ease of development is well worth watching.

Anyway, the thing is that it looks like someone has made an mbed to Arduino Shield adaptor, so connecting to various ready-made peripherals should soon be even easier.

Considering the no-hassle compiler and libraries are free, the mbed is a quite reasonable proposition for dipping your toe in the ARM embedded development waters at about the same price as the far less powerful Arduino Nano.
I got mine from Element 14 (formerly Farnell) for about AUD $87.

Sunday, October 10, 2010

Virtualisation for Beginners

I recently learnt that Dr Dobbs Journal is no more, having just become a sub-section of Network World.

This prompted me to go pick up the article code ZIPs from the old web-site to trawl for any useful bits.

I pointed my Getright Browser Tool at - and downloaded the lot (1.31GB of zip files) while it is still there.

Amongst the goodies I found that I had downloaded a zipped image for a Ubuntu Dot.Net Development virtual machine.

I put this Zip and the FREE VMPlayer from VMWare onto an old 786MB / 40GB / AGP video Pentium PC running fully up-to-date XP SP3.

Honestly, there are probably better machines hardware-wise put out for recycle at the side of the road.

I unpacked the 1GB ZIP which became about 11GB on account of a 10GB disk image in there.

I installed and ran the VMPlayer and was happy to see that it actually ran OK on such a low-spec machine and handled the wireless 11n NIC OK.

I tried to run the Ubuntu virtual machine but if failed as it was created to run as a 2-CPU machine though luckily only with 512MB RAM.

I changed the VM settings (so easy that the cat could do it) to 1 CPU and away it went.

Then VMPlayer announced that my Linux VMWare Toolkit was out of date and offered to fix it up while I was still using the Ubuntu VM.

I accepted and the tools were downloaded verified and mounted so that they suddenly appeared as a CD inside the Ubuntu VM, Impressive!

I opened the .tar.gz tools archive file on the CD and dragged the contents to my home/user directory and they were uneventfully extracted.

I then ran the enclosed .pl install script and voila, my Ubuntu virtual machine is all updated with the latest VMWare tools.

Now I can play with Ubuntu Dot.Net development any time I want without having another old PC hanging around.

Actually, I’m tempted to get VM Workstation and make a VM of the old XP machine so that I can run IT under Windows 7 in the future.

And NO I won’t even try and run Ubuntu in VMWare in the virtual XP machine, I’ll make a separate Ubuntu machine. :-)

So, two take-aways from this –
  1. Getting a hands-on taste of Virtualisation is cheap and easy, and…
  2. You don’t need a high-end machine to experiment on (and get some Virtualisation experience you can claim when job-hunting).

Have a go! :-)

Thursday, March 25, 2010

General: My Obesity Solution

Here is a passing thought I will share with you.
The media and doctors are all brow-beating us about how the population is becoming obese.
Well surely, if so many people are becoming obese, then it should be time to redraw the height/weight scales that are used to determine 'normal'.
If you think that would be cheating consider this - the scales currently in use were derived post-war after years of rationing where people barely saw sugar or butter or any sizable piece of meat, is that fair? Is it pure coincidence or artistic incompetence that paintings and books from previous centuries depict a great many portly people? (Think 'Pickwick Papers').
I think that losing weight might be a good plan, but I am not convinced that we are being judged fairly.

Others doubt the BMI too!

Update: A nice kicking for BMI on The Register -'Being fat is no worse for you than being a woman'

Friday, January 29, 2010

General: Embedded Development

Unless you are manufacturing your own line of products, embedded development starts with a customer wanting to create or improve a new or existing product.
Then follows a lot of question and answer and,l usually, research to identify the possible range of solutions, from off-the-shelf to fully-custom, and consider their respective strengths and weaknesses.
If it seems to both parties that a custom solution is the best option, you may have work to do, if the business matters are agreeable to you both.

Then, in theory, begins the thinking and design of a possible solution, although in practice quite a few possibilities will usually have occurred to the designer during the discovery phase.
The next step is fleshing out the component parts of the overall solution, which may start as no more than a few quickly drawn ideas, like so -

These rough ideas are frequently tested by the designer, either as components, or as a part of a whole solution. Whilst the older method is to build test hardware, a lot of people opt for simulation nowadays.
In order to produce the prototype, the schematic designs must be entered into the design tool and parts created and added to the library if it does not contain the required part, or not one in the required physical package.

When the complete schematic is entered and verified for electrical consistency, the board layout can be performed. An important first step here is entering the manufacturing capabilities of the PCB company as a set of design rules that your board must conform to.

When the layout is complete, a set of files can be sent to the PCB manufacturer and a small batch of prototype boards ordered. It can typically take 5-10 days to receipt of the finished boards, although speedier options exist for higher prices.

This board is mostly Surface-Mount, apart from power devices all lined up to share a common heatsink and the much cheaper through-hole electrolytic capacitors. The hardest-to-solder ARM processor goes on first to give unimpeded access to it!
Once the prototype is fully debugged, actual assembly will be performed at the PCB manufacturers, but for the prototype it's a job for a good eye and a steady hand.

Once all parts are fitted, the board can be fitted into the prototype hardware and then the software programming and debugging can commence!

In this case, a JTAG port makes software development and testing much easier.

Larger size photo's can be viewed in this Flickr set.

Saturday, January 23, 2010

General: Test your ISP!

Here is a handy site to see what shenanigans your ISP might be up to - Netalyzr

Here are my results!

Summary of Noteworthy Events
Address-based Tests +
NAT detection (?): NAT Detected
DNS-based host information (?): OK
Reachability Tests +
TCP connectivity (?): OK
UDP connectivity (?): OK
Path MTU (?): OK
Network Access Link Properties
Network latency measurements (?): Latency: 250ms Loss: 0.0%
The round-trip time (RTT) between your computer and our server is 250 msec, which is good.
We recorded no packet loss between your system and our server.
TCP connection setup latency (?): 250ms
The time it takes your computer to set up a TCP connection with our server is 250 msec, which is good.
Network background health measurement (?): no transient outages
During most of Netalyzr's execution, the applet continuously measures the state of the network in the background, looking for short outages. During testing, the applet observed no such outages.
Network bandwidth measurements (?): Upload 220 Kbit/sec, Download 1.3 Mbit/sec
Your Uplink: We measured your uplink's sending bandwidth at 220 Kbit/sec. This rate could be considered somewhat slow, and may affect your user experience if you perform large transfers.
Your Downlink: We measured your downlink's receiving bandwidth at 1.3 Mbit/sec. This level of bandwidth works well for many users.
Network buffer measurements (?): Uplink 4600 ms, Downlink 1400 ms
We estimate your uplink as having 4600 msec of buffering. This is quite high, and you may experience substantial disruption to your network performance when performing interactive tasks such as web-surfing while simultaneously conducting large uploads. With such a buffer, real-time applications such as games or audio chat can work quite poorly when conducting large uploads at the same time.
We estimate your downlink as having 1400 msec of buffering. This is quite high, and you may experience substantial disruption to your network performance when performing interactive tasks such as web-surfing while simultaneously conducting large downloads. With such a buffer, real-time applications such as games or audio chat can work quite poorly when conducting large downloads at the same time.
HTTP Tests +
Address-based HTTP proxy detection (?): OK
Header-based HTTP proxy detection (?): OK
HTTP proxy detection via malformed requests (?): OK
Filetype-based filtering (?): OK
HTTP caching behavior (?): OK
JavaScript-based tests (?): OK
DNS Tests
Restricted domain DNS lookup (?): OK
We are able to successfully lookup a name which resolves to the same IP address as our webserver. This means we are able to conduct many of the tests on your DNS server.
Unrestricted domain DNS lookup (?): OK
We are able to successfully lookup arbitrary names from within the Java applet. This means we are able to conduct all test on your DNS server.
Direct EDNS support (?): OK
EDNS-enabled requests for small responses are answered successfully.
EDNS-enabled requests for medium-sized responses are answered successfully.
EDNS-enabled requests for large responses are answered successfully.
DNS resolver address (?): OK
The IP address of your ISP's DNS Resolver is, which resolves to Additional nameservers observed for your host:
DNS resolver properties (?): Lookup latency: 200ms
Your ISP's DNS resolver requires 200 msec to conduct an external lookup.
Your resolver is using QTYPE=A for default queries.
Your resolver is not automatically performing IPv6 queries.
Your DNS resolver does not use EDNS.
Your DNS resolver can successfully accept large responses.
Your resolver does not use 0x20 randomization, but will pass names in a case-sensitive manner.
Your NAT has a built in DNS proxy. The DNS request was received from
DNS glue policy (?): OK
Your ISP's DNS resolver does not accept generic additional (glue) records — good.
Your ISP's DNS resolver does not accept additional (glue) records which correspond to nameservers.
Your ISP's DNS resolver does not follow CNAMEs.
DNS resolver port randomization (?): OK
Your ISP's DNS resolver properly randomizes its local port number.
The following graph shows DNS requests on the x-axis and the detected source ports on the y-axis.

port sequence plot

DNS lookups of popular domains (?): OK
77 of 77 popular names were resolved successfully.
7 popular names have a mild anomaly. The ownership suggested by the reverse name lookup does not match our understanding of the original name. The most likely cause is the site's use of a Content Delivery Network.
2 popular names have a mild anomaly: we are unable to find a reverse name associated with the IP address provided by your ISP's DNS server. This is most likely due to a slow responding DNS server or misconfiguration on the part of the domain owner.
DNS external proxy (?): OK
Your host ignores external DNS requests.
DNS results wildcarding (?): Warning

Your ISP's DNS server returns IP addresses even for domain names which should not resolve. Instead of an error, the DNS server returns an address of, which resolves to You can inspect the resulting HTML content here.

There are several possible explanations for this behavior. The most likely cause is that the ISP is attempting to to profit from customer's typos by presenting advertisements in response to bad requests, but it could also be due to an error or misconfiguration in the DNS server.

The big problem with this behavior is that it can potentially break any network application which relies on DNS properly returning an error when a name does not exist.

The following lists your DNS server's behavior in more detail.

  • www.{random}.com is mapped to
  • www.{random}.org is mapped to
  • fubar.{random}.com is correctly reported as an error.
  • [sic] is mapped to
  • nxdomain.{random} is correctly reported as an error.
Host Properties
System clock accuracy (?): Warning
Your computer's clock is 14 seconds slow.
Browser properties (?): OK
The following parameters are sent by your web browser to all web sites you visit:
  • User Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
  • Accept: text/html,application/xhtml+xml,application/xml; q=0.9,*/*; q=0.8
  • Accept Language: en-us,en;q=0.5
  • Accept Encoding: gzip,deflate
  • Accept Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Java identifies your operating system as Windows Vista.
Uploaded Data (?): OK
The following additional data was uploaded by the applet: