Friday, March 26, 2010
Another tune. This one’s better [Updated]
No, honest, I think this one’s better. I had help: Martin played some of the sax and the fragrant Becky sang some backing vocals.
My vocals are the weakest thing about this song; now I’m getting the hang of recording, I really need to find some decent lead vocalists to record :) I’m happy with the song itself, though. Nice funky bits.
[Updated: 26/3/10 - corrected the drums dropping out in v2]
Tuesday, March 23, 2010
Music: Never worked before
I bought a new mic to use for voice-over work - still at the cheap end of things, but a gazillion times better than the old Radio Shack highball I used to use. Mmm. Shiny.
Thought I’d have a quick go at recording a song with it. I’m no great vocalist (read: earplugs please) but this song marks several firsts for me:
- First time I’ve finished a song (well, ok, first time I’ve got as far as 3 verses and fade; production-wise it probably wants redoing from the ground up by someone else)
- First time I’ve written a tune without any percussion
- First time I’ve written lyrics that didn’t make me puke as soon as I heard it back
Vocals start really bad and then improve swiftly to become merely mediocre. It’s pretty soft and wussy, but I’m quite pleased with it. Harmonies are fu-u-u-un to do.
iPad thoughts…
Gotta get me one of those shiny new tablets…
Where I’m coming from
I’m a reluctant Mac switcher, having been forced to buy one by a couple of editing-heavy video jobs last year. To make life easier I shifted as much of my data as possible into the cloud: Gmail, Google Docs, Xmarks (for syncing bookmarks). As time has gone on I’ve found I use the PC less and less - only really for games and 3D software.
OS X is not better than Windows, it’s just a different set of problems. I manage to tangle my Mac up badly enough to need to reboot just as often as I did with the PC. There are some nice UI touches, along with a host of stupid Mac-specific UI gotchas that irritate the hell out of me. Some of them hint at Apple’s stubbornness - across the OS, you can only resize windows by dragging one little corner of the window, but yet in the Apple Pro apps, Logic, Final Cut Pro etc, they’ve implemented a drag-any-border-to-resize facility, as if they *know* that’s how it should work. They know it’s right, but they don’t implement it across the OS because it’s one of those Apple UI things that’s been around since the beginning of time.
Well, there you go: Windows and OS X are kinda different, but mostly the same. The thing that holds them both back:
I am Legacy
Legacy support is the one thing that cripples technology development. There are three layers of legacy needs that tie us to the past: the two obvious ones, hardware (seen how many drivers Windows 7 carries along, just in case you decide you want to use a dusty old PS/2 mouse?) and software (because no one wants to tell you you have to buy new versions of all your applications when you upgrade your OS, and yet we’ll happily do it if the application itself purports to give us new features). The third legacy tie is the user interface paradigm. We’ve spent years learning the ins and outs of our computers. We’ve learnt all about files and folders, windows, clicking, dragging, control-c and v; we’ve learnt how we need to open an application before we can write a letter, how we have to “save” it or it’ll disappear when we switch the computer off, and so on.
This last tie is the toughest thing we’re being asked to give up by Apple. Hardware and software are just money, but the learning that we’ve done, the skills we’ve learnt - now *that* was real effort on our part, and to accept a whole new way of computing means letting go of those skills that set us apart from the great unwashed masses. If I’m honest with myself, I have to recognise that I have an enormous pride in my computer skillz; I’ve learnt how to tame the digital beast. People come to me for help because I am wise and learned.
But I am Legacy. The future is not a place where people will need experts like me to help them do simple things like printing a letter.
Bring Copy’n’Paste, leave the rest
The iPhone was the perfect way for Apple to take a bold step towards a new future. We’d baulk at learning how to use a brand new computer operating system, but somehow we don’t perceive the iPhone as a computer - it’s an appliance, and we don’t mind learning how to use new appliances because they don’t threaten our pedestal of *real* computer knowledge.
This gave Apple a way to start the revolution on foreign soil, as it were - they could ditch the legacy stuff completely without anyone really minding. New hardware approach (touch, plus one single hard button; no mouse or keyboard), new software, and a whole new interface paradigm. No more files and folders, windows, pointers. Everything’s geared around the tasks: make a call, update your calendar, answer an email.
The only thing that seems remotely familiar from the “real computers” is copy and paste - interestingly, it was the one feature of my Dad’s word processor that impressed him the most when he finally moved over from the typewriter - but there are stories there for another day.
Bring it on
So the thing that’s irritated the Tech community the most about the iPad seems to me to be this: Apple are showing us that *we* are the legacy. We generally like shiny new hardware, and we like clever new software, but giving us a whole new simpler computing paradigm is basically telling us that the skills we’ve been developing for the last umpteen years are now obsolete. It’s like throwing the shepherds in with the sheep and saying “you are all equal now”.
It seems that at least some of the scorn the iPad has received from the tech community stem from this root. Not all, obviously - there are some genuine concerns about the App Store and its approval process, but for now it’s the least worst solution.
Well, a few short thoughts have turned into a whole long page, so I’ll leave it there for now. Next time: multi-tasking, hacking, and the App Store.
Sunday, March 21, 2010
LED lighting
Joel at Design & Make has the right idea.
The problem with white LEDs
The white LEDs that have flooded our lives these days aren’t really very white. They’re made up of a bright blue LED with a blob of phosphor on the top that converts a portion of the blue light into an orangey color. Together, the two colours merge to look kinda white to our eyes, but because the light is missing lots of the wavelengths of light that normally make up daylight, coloured objects often look oddly murky under it.
The solution is to combine more wavelengths of light to create our white lights. Red, green and blue are the usual combination - in the right proportions, you can make a light of almost any visible colour. Get the balance right and you can make a decent white light that mimics daylight and makes the coloured objects it falls on look the right colour.
It’s more complicated to set up than just using the pseudo-white LEDs though. So for a lazy approach to non-critical accent lighting, white LEDs will do. Just to make it even easier, you can now buy pre-wired white LEDs on a roll from eBay - this roll cost £37 for 5 metres, which is pretty damn good for lighting that’ll never burn out.
You can cut it up, it runs from 12v DC, and it comes sheathed in a nice protective silicon sleeve. It’s lightweight enough so you can just use double-sided sticky tape to stick it up, too: my fave is under desks:
OK, so it’s a bit overkill, but I hate overhead lighting.
Wednesday, June 17, 2009
New speedo for the Landrover
My speedo cable snapped. Boo. The speedo was never very accurate anyway, so rather than replace the cable I figured it’d be fun to try making a digital speedo - plus it’s an opportunity to try using surface mount components rather than the bigger through-hole ones I’ve used so far.
A bit of an ambitious design this time round - separate display and control boards, bigger and more complicated than anything I’ve done to date:
First - hacking up bit of unetched PCB. You get through a lot of stanley knives like this.
Exposing the board:
Etching it:
The tracks on it are tiny this time around:
Ended up having to redesign part of the board as the pads were too small for me to drill holes through. I need a better drill. Still, good practice for tring to solder some of the tiny SMD components on:
New design, etched and with all the components on. SMD components are actually much quicker and easier to solder than through-hole, it turns out. The board’s a bit messy ‘cos some of the tracks broke and I had to solder little strands of wire across to fix them, but still:
With the control board finished, it was time to connect it to the display board (which will house the actual LED displays). To connect it up meant over 50 tiny little wires had to be stripped and soldered to each board:
Finally, all the wires are soldered up - the boards are connected …
… this is what it looks like from the back:
The displays plug into the sockets on the display board:
The boards fold together, using all those little wires as a hinge. I made a simple Meccano frame to hold the boards (rather than create a neat box to put the whole thing in, I quite like having it all guts-out)
A little wire loop holds the bottom of the boards together so they don’t flap around:
Done:
Then a couple of magnets holds the thing to the windscreen frame on my truck. Magnets rule.
And this is the final thing - in position, working. Normally it shows the speed, but when it detects the Landrover hasn’t moved for 5 seconds it alternates between showing the trip distance and my maximum speed on the trip:
And the best thing is that it’s a darn sight more accurate than almost all dial-based speedos, especially factory fitted ones (most car speedos read 4-5mph faster than you’re actually going, to err on the side of caution - when your car speedo says you’re doing 30mph, you’re probably only going 25-26mph. The manufacturers figure it’s safer to have you going slower than you think rather than the other way around - compare your speedo with your satnav’s speed readout if you have one).
*** UPDATE ***
Had a few questions about this thing by email, so a few more details:
- The thing runs off an Atmega328. That’s what’s so cool about Arduinos - once you’ve got your circuit and code working, you can take the programmed chip out of the Arduino board and solder it into a stripboard with a crystal, 2 capacitors and 2 resistors and it’ll carry on working. Then you can stick a fresh chip in the Arduino (at a cost of about £1.50) and get on with the next project.
- The sensor is an Allegro Hall Effect sensor, which gets triggered by a tiny magnet stuck to the propshaft. The time between the pulses is inversely proportional to the speed you’re going, so it’s a quick simple calculation for the Arduino.
- The left-most display is an HDSP-2113 smart alphanumeric display. Takes a good few data lines to drive all the displays, hence the use of all the shift registers (74HC595s) on the rear PCB. Shift registers ru-u-u-ule.