Andrew's Stuff

LED Light Bulb Comparison

Comments

Background

As part of our attempts to lower our currently-£180/month electricity bill, we've been looking at replacing some of our light bulbs with lower-Wattage equivalents. Despite only being built 12 months ago, this house was fitted with some pretty high-energy lighting. As per the NHBC requirements (I believe), the hallways were all fitted with annoying 4-pin "energy-saving" sockets, but the other rooms all came with incandescent bulbs! Not bad, given that the UK banned such bulbs only nine months later! Most of the rooms had standard bayonet cap fittings, so we replaced those immediately with cheap-but-perfectly-adequate CFLs. However the kitchen and the three bathrooms all had Halogen GU10 sockets, each fitted with a 35 Watt Halogen bulb. With eight in the kitchen, two each in two of the bathrooms and three in the third, that amounted to fifteen 35W bulbs for a total power draw of over 500 Watts! The kitchen alone uses 280 W when you turn the lights on. The ones in the kitchen aren't even positioned over the work surfaces or sink, making them near-useless, but that's a rant for another day.

Anyway, since LEDs are the new, up-and-coming light source of the future we decided to investigate our range of options in that area. I had already purchased a GU10 LED bulb from Ikea the best part of a year ago, and a GU10-fitting CFL from Tesco at about the same time, but neither were much good so we hoped that some other options would be more reasonable. As such, we went to Homebase and Sainsbury's and bought whatever we could find that looked remotely useful. We had already decided that lighting all of the kitchen wasn't necessarily useful, and that, therefore, most of the kitchen would suffice with lower-brightness bulbs, with just a couple of high-brightness ones near where we're actually preparing food. This is why our purchase included a pathetically-low-brightness 1 Watt LED bulb.

Dell Studio 1735 Laptop

Comments

I recently purchased a new laptop - a Dell Studio 1735. Its price tag of £1250 is high, but acceptable given the specification and options I added:
  • 2.1GHz T8100 Core2Duo CPU
  • 3GB RAM
  • 320GB 5400 rpm hard disk
  • ATI/AMD Radeon HD Mobility 3650 256MB
  • 17" 1920x1200 LCD
  • Windows Vista Ultimate
  • 802.11n WiFi card
  • Bluetooth
  • Gigabit Ethernet
  • DVD+/-RW drive
  • Integrated webcam
  • Backlit keyboard
  • Fingerprint reader
  • Multi-card reader (SD, MMC, MS, etc.)
  • Logitech VX Nano mouse (wireless laser mouse with Logitech's "nano" receiver that barely protrudes from a USB port)
  • 4-year warranty
  • Also possesses an HDMI output


2787542475_775fedaa82

UTF-8 Support

Comments

After spending ages working with UTF-8 the other day, as I mentioned in my previous post, I've finally added UTF-8 support to this site. Both comments and posts by me are now able to contain all kinds of crazy characters without resorting to complex HTML entities!

Examples:

UTF-8 in PHP and MySQL

Comments

I spent the best part of today researching how to properly handle Unicode - specifically UTF-8 - with PHP and MySQL for Web applications. Since this is something that has stumped me for a while, I decided to make this post about it. This will be as much for my own benefit as anyone else's: hopefully I'll remember it for future use if I write about it, and I can always refer to this post at a later date if necessary.

A quick and very basic introduction to UTF-8: whereas the ASCII character encoding that we're all taught about first at school stores characters very simply with one byte per character, UTF-8 uses anything from one to six bytes to store each character, depending on how many bytes are actually needed to store it. This means it can store many more characters than ASCII or any other one-byte-per-character encoding which are limited to a maximum of 256 "characters" (the first 32 ASCII codes are actually unprintable control codes, leaving a mere 224 values for actual characters), which is enough for English (mostly) but not enough for the vast majority of other languages which have accents and all kinds of fancy letters. And that's before we get onto Japanese, Chinese, Korean and the like with thousands of characters each. UTF-16 and UTF-32 also allow many more than 256 characters to be stored, but they take 16 bits and 32 bits, respectively, per character to store data, which is fair enough if you're using a lot of high-numbered characters, but a huge waste of storage space, bandwidth and/or memory if you're primarily using the basic ASCII characters which otherwise fit in a single byte each. UTF-8 gets around this by storing the lower ASCII values (i.e. just 0-127 - the original ASCII standard, rather than any of the extended ASCII sets which filled in the remaining 128-255 range) in a single byte, and using the most significant bit to indicate if another byte is needed to store the character, allowing the number of bytes used by a character to grow if necessary, but only if it actually is necessary. This also has the pleasant side-effect of keeping the majority of the English language characters in the same position as they are in ASCII: only the odd characters like £ are moved around.

My Xbox 360 Died Again!

Comments

Update (2007/09/22):
I should probably update this to say that on the 30th August I received a refurbished Xbox 360 from Microsoft that not only works (no problems thus far, and I've had many a 5-hour gaming session on it), but it also considerably quieter than my last one. So it would seem that contrary to what I had read elsewhere, Microsoft are still sending refurbished units out instead of fixing and returning people's original console. For me, this was a good thing, but for others I can imagine it wouldn't be.


Well, here we go again.