Expand Cut Tags

No cut tags
ggreig: (Western gentleman)

God help us! The Iguanodon's loose!

A tranquil street scene – just before the populace become aware that an Iguanodon has escaped from the Zoological Gardens.

There’s a Highland company called Antediluvian Miniatures that have started producing proper miniatures of dinosaurs, taking into account the very latest scientific thinking – of the 1850s. As yet their range is small, but includes the two most iconic early representations of dinosaurs: the Crystal Palace Iguanodon and Megalosaurus! (Also, not included in this post, but I have to mention them – three intrepid adventurer figures including Shug McClure, Raquel Scotch and the finest of all: Professor Peter Cushion, adjusting his monocle and preparing to fend SOMETHING off with a furled umbrella.)

I should also mention that Antediluvian Miniatures have a very cool t-shirt, featuring their mascot Professor Buckland.

The real Crystal Palace IguanodonsThere’s a good chance you’re aware of the Crystal Palace dinosaurs, if not, then Wikipedia and other places are your friend.

The Iguanodon figure isn’t a perfect replica of either of the ones at Crystal Palace, but it’s more like the one standing upright.

The Crystal Palace dinosaurs are often given as an example of how scientists of an older generation got things hilariously wrong, especially the Iguanodon with the horn on its nose (now known to have been a thumb-spike), but the Iguanodons actually show a greater humility from Sir Richard Owen than our caricature allows. The two Iguanodons are different, with the one that Antediluvian have taken as their inspiration standing upright, while the other one is more lizard-like, and lounges on the ground with one paw up on a tree-trunk. There was doubt even at the time that these reconstructions were correct – they were just the latest theory.

When painting these two I tried to get something in between the look of the statues and something that could be a real beast, so the Iguanodon is a bit more vibrant than one of my paint jobs would usually be, making the faded shade of the statue look more lively. That’s the current colour of the statue, of course, as that's what I could find in photos; the colour they’re painted has changed over time as well as our theories of what the beasts were actually like.

Iguanodon wandering the Zoological Gardens

I defy you to spot the joins – both the Iguanodon and the Megalosaurus come with separate legs. I did apply a bit of Milliput as filler, but the fit of the moulded parts was really good, to a level that I know must be difficult for figure designers to achieve, judging by the frequency with which they don’t attain it. I was really impressed with these models. The Iguanodon is resin with metal legs, while the Megalosaurus is all resin.

Iguanodon figure inspired by Crystal Palace - right sideIguanodon figure inspired by Crystal Palace - left side

Rather annoyingly, there’s a mould-line that shows up in these photos of the Megalosaurus that’s actually hard to pick up with the naked eye under most conditions. The light in these photos hit it just right – or wrong. It’s also intended to look like a potentially living version of the real statue. The Megalosaurus also comes with scale replicas of the original fossils (not included in these pictures, and not yet painted, though I have some other scale fossils for them to go with).

Megalosaurus figure inspired by Crystal Palace - right sideMegalosaurus figure inspired by Crystal Palace - left side

ggreig: (Western gentleman)

Or, Reinventing The Torch.

In Doctor Who, the Master’s signature weapon is the Tissue Compression Eliminator, which kills by shrinking the target to the size of a doll (the scale of the doll is, for some reason, never specified). It’s mostly used by the Master as played by Anthony Ainley in the 1980s, and in those stories takes the form of a thick rod with a bulb at the end that opens, crocus-like, to fire a red beam.

However, it made its first appearance in the Master’s very first story, Terror of the Autons, in which Roger Delgado’s Master wielded a rather more stylish version resembling a cigar, and fired by pressing with the thumb:

The Ainley period Tissue Compression EliminatorThe Delgado period Tissue Compression Eliminator

Knowing I was going to be attending a Doctor Who convention and would be expected to dress accordingly, it occurred to me – before last season’s revelation about the Master – that a) I had the beard for the part and b) had always fancied a jacket with a Nehru collar. A bit of hair colouring would convert my largely white beard to the original Master’s two-tone badger-style goatee. Couldn’t do much about the hairline.

What I couldn’t easily accept, adapt or get hold of, however, was the Master’s TCE, so I thought about building one.

The Core

My first thought was to buy aluminium tubing and telescope short sections of it together, but it proved difficult to find sizes guaranteed to telescope and although the prices were reasonable for the amount I wanted, minimum order amounts were – less so. I headed out to the nearest hardware store and inspiration, of a sort, struck. Toilet roll holders – the sprung, plastic sort – would do the job for a few quid.

I bought a handful of black ones on the spot to experiment with, and ordered some silver ones from eBay. The black ones were noticeably better quality, but using a silver one as my base would give me a much better finish for the sliding section than I could possibly have achieved using paint. I also found I could fix some if its issues by opening up a big enough hole in the larger end to work the weedy spring out and replace it with a stronger spring from one of the black ones.

The Circuitry

Having found something that could form a good base to work on, the next thing to consider was what to put inside it if I wanted it to appear to work. Well, electrically that’s not hard. What I would need would be a power source, and something that would light up when a switch was closed. It’s a torch!

Circuit diagram for the Tissue Compression Eliminator

With appropriate circuit board to mount them on, I could fit seven LEDs (arranged hexagonally, with one in the middle) within the diameter of the toilet roll holder. A bit of browsing at Proto-Pic turned up what I wanted; a 1" circular protoboard. I did have to trim it down a bit to fit, but it gave me a good starting point. I also got some Super Bright Red LEDs. I already had resistors to hand, and a small selection of switches which happened to include a suitable one.

The circuit couldn’t get much simpler and was easy to throw together on a solderless breadboard just to check everything worked. Then the tricky bit was to work out how to wire it up in practice, cram it all inside a toilet roll holder without breaking anything, and provide a way of activating the switch.

Building The Shell

It was obvious the LEDs would have to be mounted inside and at the front somehow. However, a single toilet roll holder was neither long enough to represent the TCE, nor would it be possible to mount the LEDs inside its broader half – there would be nowhere to attach them, and that space also contained the spring. Even if that hadn’t been the case, it would have been very tight and might have necessitated cutting down on LEDs. This is where the black toilet roll holders came into play.

By cutting the bigger part of a black toilet roll holder to a suitable length, and then sawing a slit along it, it was possible to prise it open and fit it as a sleeve around the silver one. This again gave the right colour without the need for painting, and it made it possible to extend the length of the TCE with a double skin – an inner, un-slit tube simply stacked on the front of the silver one, and a slit sleeve embracing them both. The slit tubes would require filling later to cover the gap, but again a good structural foundation was in place.

With the idea for the extension of the muzzle established, the problem of where to mount the LEDs was solved, as they would go inside that rather than the body of the main holder.

Planning The Interior

With that settled, the next most obvious decision was that power would have to go at the other end, so that it could be easily inserted and replaced. To power the LEDs, I wanted a 6V power supply. Within the space available, that meant I’d need to go with with several watch batteries, but I managed to get an AAA battery holder that fit nicely inside the thinner half of the toilet roll holder, and found that LR44 watch batteries would fit nicely within that. Taping four 1.5V LR44s into one slightly bigger 6V battery with insulation tape and adding a longer spring to the battery holder to hold them in place solved the power supply problem.

With both power and light source at least notionally sorted out, the tricky bit was how to connect them up successfully, with a switch in between that would be activated by the narrow end of the toilet-roll holder being pushed in.

How To Press The Switch

I had a small press-to-close switch that would fit inside the narrow end, and was about the same width as the battery holder. I could mount it (and the resistor) at the positive end of the battery holder, and hopefully attach some sort of plunger to the far end of the toilet-roll holder that would come down and press the switch closed when the smaller tube was pushed into the larger one.

I considered other alternatives, the main one being conductive paint on the outside of the small tube being brought into contact with more on the larger tube, but whenever I thought about reliability, I came back to the plunger option. Paint would have been very prone to wear and tear, and (as the conductive paint is black) painting over it to restore the silver look would reintroduce connectivity problems and spoil the look.

However, a plunger presented problems too. It had to be long enough to press the switch, but not so long that it would cause mechanical problems by pressing too hard – a tricky measurement to make confidently inside a tube and out of sight. It had to be broad enough to be sure of hitting the switch, and to prevent it being easily knocked out of alignment, but narrow enough to avoid catching on the lip of the narrower tube, and getting caught up in the spring. Finally, it had to share the confined space with the wires connecting the power end of the assembly to the LEDs, without fouling or putting strain on them – or doing the same to itself.

The solution I came up with was to build a plunger that was fixed at one end, but had the other end sprung.

Building The Plunger

In order to accommodate a spring with the least hassle possible, I started with an old modelling paintbrush for the shaft of the plunger. It was roughly equivalent to a piece of narrow dowel, but with the advantage of being already hipped. I measured things out and cut it so that I could just slide a small spring on, and the broad part of the paintbrush would hold it in place. Then I built up each end with Sugru and a circle of plastic card to fix things, provide flat surfaces at either end, and make the plunger robust.

Making one end of the plunger sprung provided a couple of benefits. It allowed for a soft press on the switch, so that it didn’t matter so much if I didn’t get my measurements quite right; and it prevented catching on the lip of the narrower tube from being a problem – if it did occur, the sprung tip would flex slightly until it just slipped in, rather than getting stuck.

The most worrisome part was getting the wiring to share the same space. Ultimately, I just drilled holes through the ends of the plunger for each wire, with enough space to let it move fairly freely, and hoped.


Parts before assembly - click through for full sizeHaving – more or less – worked  out all the parts necessary, it was time to try to bring everything together. I’ve not done a lot of soldering in the last 25 years, so creating a circular array of 7 LEDs wired in parallel kept me busy for a while, and worried about short circuits, as there wasn’t a lot of space to work with the wire I had. Connecting up switch and resistor on a little piece of protoboard and connecting it to both the battery holder and the wire that would connect to the LEDs was challenging within the space available too. Both of these went well, though, and I used more Sugru to fix the switch assembly firmly in place (and provide a bit of additional insulation) at the positive end of the battery holder.

Where I messed up – though I didn’t discover it until later – was the simplest part of the soldering. While connecting a wire to the negative terminal of the battery holder, I overheated the plastic and managed to break the connection between the metal terminal and the spring holding the battery in place. I tried to keep testing everything was still working at every stage, but I missed this and had some worrying debugging of connections later on with a multi-meter when stuff just didn’t work! When eventually discovered, a combination of conductive thread and paint remedied the issue.

I fed the wires through the plunger, with excess that could be cut back once I knew how much slack would be required to attach the LED assembly but allow the battery holder to be pulled out for inserting or removing batteries, and I drilled some extra holes through the stationary end of the plunger to allow me to fix it in place more firmly with thread or wire. I tried thread first, but it was too fiddly and difficult to get right, so I fell back on aluminium modelling wire – and another layer of Sugru to fix the LED assembly in place and avoid short-circuits!

TCE nearly complete

Finishing The Exterior

As already described, the exterior of the barrel was to be built up with parts of the black toilet roll holders. With a bit of specialist super glue for awkward plastics, these went on firmly and the large lengthwise gaps left by slitting them and the smaller circumferential one between the two sections were filled with black Sugru, which I then rolled on greaseproof paper to try to get a reasonably smooth surface. The result wasn’t perfect, but good enough if you weren’t inspecting it closely. The interior of the barrel was painted silver.

The butt end, where the battery holder was inserted, was covered up with a cylindrical rubber ferrule (sold for the foot of a chair), which was just right for the job of a battery cap.

Finally, I added a a bit of copper trim – two strips of plastic card covered with Bare-Metal Foil and attached with the plastics super glue. One covered the join between the two sections of black tubing nicely.

The End Result

The completed TCEThe completed TCEThe completed TCE, lit upThe original, on-screen TCE

The final picture shows the original, for comparison.

ggreig: (Western gentleman)

Interesting story on the BBC web site, with video, about a serious attempt at a modern airship, hybridised with a flying wing. (Its airship ancestry is more obvious.) You may have seen pictures before, as it’s a project that’s been resurrected in a new form, but interesting to see it within a hop, skip and a jump of becoming a commercial reality. May contain trace quantities of Iron Maiden.

Dino 101

Jan. 19th, 2014 05:09 pm
ggreig: (Western gentleman)

dinologoLast week I got my certificate for completing Dino 101: Dinosaur Paleobiology, from the University of Alberta. Like most people, the idea of dinosaurs gripped me when I was wee. I decided before getting to primary school that I wanted to be a palæontologist, and could bamboozle senior relatives with the word. (Or they were playing up to me. I couldn’t tell at the time and prefer the first version now!)

Time went by, and when I got to secondary school, I suddenly twigged that if I wanted to become a palæontologist, I'd probably have to do Biology, and that would mean cutting things up. A combination of squeamishness and conscience meant I really didn’t want to do that, so I gave up on my dream of palæontology and switched to wanting to work in electronics (also not right for me, but it took more years and a degree to reach that conclusion).

Last summer, a story on the BBC web site alerted me to Dino 101, a MOOC (Massive Open Online Course) starting in September, and I signed up.

Having been casually interested in dinosaurs for many years, I already knew quite a lot of material that the course covered – it is introductory, after all – but I also learned things. In the very first lesson, I learned what gastralia are, a detail of dinosaur (and crocodile) anatomy that despite all my interest had utterly passed me by.

Internal artwork from 'Dinosaurs - Hunter and Hunted', a primary school projectThe course is nominally a video lecture a week, over twelve weeks, with a five question quiz after the lecture.  There are also lecture notes that are worth reading as there’s some material in them that may not be covered in the lectures. You can go at your own pace, however, so some people had finished the course within a day or so of it being opened up. I took it at a lecture a week at first, but picked up pace so that I rattled through the latter part of the course and finished in early October.

The lecture titles were: Appearance and Anatomy; Eating; Moving Around; Birth, Growth and Reproduction; Attack and Defense; Death and Fossilization; What is a Species?; Evolution; Stratigraphy and Geologic Time; Paleography and Plate Tectonics; Dinosaur Origins; and Dinosaur Extinction.

The only one that I felt was not entirely successful was the lecture covering Stratigraphy and Geologic Time. It was more or less a recital of geological time periods with a small amount of detail about what characterised each one before moving on to the next. It’s difficult to think of a different way of doing this, but it didn’t make for the most thrilling or memorable of lectures. Having said that, it did hold my attention a little more than the same subject does in writing, so maybe with repeated viewing some of it will stick.

And repeated viewing is an option! Although you can watch the lectures online and get the benefit of some interactive bits, you also have the option of downloading them for offline viewing, and I did both. The lectures are presented by Betsy Kruk, a research student, rather than the nominal course-giver, Philip J. Currie. This is kind of a TV approach to lecturing, which I have mixed feelings about, ultimately but it probably was a good idea. Though Professor Currie is a recognised authority you may have seen on other programmes and a good speaker, Betsy’s better at projecting her enthusiasm on camera and, having been in quite a number of dry lectures over the years on other topics, that really does help.

The Early World;  jotter cover from a 1978 primary school project (age 10)The interactive bits are a bit of fun. Most are quick quiz questions (What do you think is right? A, B, C or D?) before giving the right answer, but there are also a few puzzles (reconstruct this dinosaur skeleton;lay out this phylogenetic tree of the families of dinosauria) and a fossil cabinet. The fossil cabinet is treated as though it were a highlight of the course, and makes appearances throughout, with new fossils added each time, but I’m afraid it is a bit disappointing. The fossils, which you can turn around and look at closely, are low resolution 3D models, with no texture-mapping – so they look like poor plastic mouldings. If they were high definition, with use made of colour to show their real appearance (possibly with false colour layers to highlight points of interest) then this could really be a highlight. Unfortunately, as it stands, although it holds some interest, a highlight it is not.

As far as qualifications go, honestly, I can’t take any real pride in my 100% record on the course, as there’s a chance to retake each quiz and I would have dropped a few questions without that option. I paid $69 to have my identity confirmed while taking the quizzes (by photograph and recognising my typing pattern while entering a declaration), but that was pretty much for vanity – I wanted some proof I’d done the course, in the form of my certificate, but wasn’t pursuing it as a genuine qualification.

If I’d wanted genuine course credits for it, I could have paid a bit more in the way of tuition fees ($263) and taken two proctored exams. The exams would have been a bit more challenging and traditional, as they’re overseen by an individual who can view your screen and see you through a web cam, so there’s no cheating. Although I didn’t take part in that, it was interesting to read how technology was being used to enable remote exams.

I would recommend repeats of this course for those with an interest, although it might be a bit simple for those who are already well informed on this subject. You can sign up here. The first repeat started on 6th January, and you might be able to join it.

I enjoyed my first MOOC, and would do another, though with limited free time I’ll have to be a bit careful what I sign up for.A more recent effort; a Natural History Museum Liopleuridon, painted in Walking with Dinosaurs colours

ggreig: (Western gentleman)

Channel 4 is doing an evening of programmes about psychopaths, and have a handy-dandy test where you can assess your own level of psychopathy.

Uninterestingly but perhaps reassuringly, it turns out I’m a teddy bear, scoring 21%:

You are warm and empathic with a heightened awareness of social responsibility and a strong sense of conscience. You like to carefully weigh up the pros and cons of a situation before you act and are generally averse to taking risks. You are very much a ‘people person’ and dislike conflict. ‘Do unto others…’ are your watchwords. But, although you avoid hurting others, those residing at the higher end of the psychopathic spectrum might not be as considerate, so stay vigilant to avoid being hurt unnecessarily.

ggreig: (I Need Dis)

A Harvard University human motor systems test that attempts to guess your age knocked 16 years off me today!

The test is based on how quickly you can click on dots on the screen – it’s a bit like that eye-test where spots of light are shone on the inside of a hemisphere and you have to react to them. I wouldn’t say my reactions are particularly good, so why did I do so well?

I think it’s because I was using a trackball. The basic assumption of the test is that you’re using either a mouse or a trackpad, although it does provide an “other” option as part of a survey after the test’s completed.

I switched from mice to trackballs a number of years ago – not sure how many, but long enough to have had to replace one. People visiting my desk hate using the trackball (so much so that I have a “guest mouse” for when people come by), and adjusting to it was hard. In fact, for the first week, it was physically painful as I got accustomed to an entirely new set of movements. After that first week though, it was full speed ahead and I’ve never looked back. A trackball’s quicker and more accurate than a mouse, and my burgeoning RSI from dragging a mouse around went away.

My trackballs are thumb-driven ones from Logitech; for some reason finger trackballs seem to be more popular, but I wouldn’t have thought they can be so quick and efficient and surely must suffer from some of the problems of mice. The thumb’s a “spare” digit from a mousing point of view, but it’s great for driving a trackball; strong, and capable of fast, sharp and quite precise movements.

If you don’t believe me, give the test a go and see how you do!


Feb. 26th, 2013 12:35 am
ggreig: (Astronaut)

I spent the first few weeks of January in California. Because I was particularly keen, [livejournal.com profile] msinvisfem and I caught a train into Los Angeles to go and visit this place:

Samuel Oschin Pavilion

We paid more than we needed to for the train, as I made the mistake of going to a window for service in the station rather than buying from a machine. Different train company, as we figured out later, and happy to charge us getting on for twice as much without even mentioning that there was a cheaper and more frequent alternative service.

We got off the train in LA Union Station, found our way outside through the impressive waiting room and walked down the road a bit for the Metro Silver Line (a bus service, despite what it may sound like). We both checked the route signage, waited for a bit, and got onto the right bus heading in the wrong direction.

When we reached the end of the route without spotting our destination, it became clear that something had gone wrong, and a helpful cop who managed to keep a straight face throughout directed us to the correct stance to go back again.

By this time, the generous safety margin that we’d allowed for getting there before our timed tickets were due to take effect was looking a bit shaky. In fact it was worse than that, as it turned out. By the time we’d returned to Go, did not collect £200, and headed out in the right direction for a similar period of time, our safety margin was completely blown and we turned up half an hour too late.

The End...eavour )
ggreig: (Default)

Sometime this month, I think I passed the milestone of 20 years of owning a PC.

I wasn’t particularly exposed to computers while I was school, nor even much at university. A friend had a BBC micro, on which I played Elite a few times, and I learned to write S-algol on the University VAXen. The closest I came to a PC was one Physics lab that required me to write a little Basic to control a stepper motor.

Then I did a conversion M.Sc. to Computer Science, and then I was unemployed.

After nearly a year of unemployment, the bright future of work as a programmer I’d hoped for looked like as though it might be escaping from me.

I thought I’d better do something about it, and with a loan from my parents and advice from my friends I bought a PC, to demonstrate that I was serious about these computer things. In particular I was grateful for the advice of [livejournal.com profile] flybynightpress, a Mac enthusiast himself, who advised me not to get a Mac but a PC, because the employment prospects were better. You tend to pay attention to advice like that!

That PC was a Viglen (pre-Lord Sugar), with a 386DX processor, 4MB of RAM, and a 125MB hard drive, with the (then) brand new Windows 3.1.

Sometime in June or July, I had a horrible experience when it stopped working. Completely. My first computer, bought with money I didn’t have, bricked. and with that horrible paranoia that maybe, oh no, maybe it was all my fault!

A technician had to come the hundred and something miles from Edinburgh to the glen where I was living with my parents on the west coast, with me liable for the several-hundred-pound cost of the callout if it really was my fault. That was not a happy wait.

Luckily, the motherboard had died of its own volition, and with a replacement fitted free of charge, that PC served me well for a fair while afterward. And a month or two later, I was employed as a programmer, albeit in a job paying only 65% of the average graduate starting salary of the time, and with the most horrible computer language in the world ever. And on a PDP-11/83, not a PC. Still, I’m sure my PC-ownership helped ;-).

Today, my Raspberry Pi arrived, both more and less powerful than that machine of twenty years ago, depending on how you look at it. Unfortunately a painful ear infection is making the thought of tinkering unappealing for me tonight; but hopefully the Raspberry Pi will be the gateway for others to follow in the career that the PC has let me enjoy.

ggreig: (Default)

Apparently the sonic screwdriver has been developed right here in Dundee! (Usual hyperbole applies, but it’s still an interesting development.)


Apr. 1st, 2012 10:02 pm
ggreig: (Western gentleman)

An Arduino UNO and accessories

Back in 1990 I was relieved to graduate with a 2:2 in Physics and Electronics, and the knowledge that whatever I was going to do next, I didn’t want it to involve Physics. A conversion M.Sc. to Computer Science came along, and having enjoyed the logical parts of Electronics, it turned out I was able to make a go of computing.

I think that probably the last time I had anything much to do with hands-on electronics was back then, when I created a voice-delay board for my Honours project; a project that would store up to eight seconds of speech at 8-bit quality, before playing it back.

For the last year or so, I’ve been intrigued by the idea of the Arduino (or a clone such as the Netduino, which allows you to use the .NET Micro Framework), a microprocessor with an environment of plug-in accessories that’s intended to make it easier for people – especially artists and designers – to achieve interesting things with electronics.

For this week off, I dug out the gas soldering iron I’ve had pretty much unused for ages, ordered a bunch of interesting sounding parts and, driven by visions of having a robot zipping about the place autonomously and chatting to people, I set out to see what I could achieve.

To avoid disappointment straight away, you possibly won’t be astonished to hear that there’s nothing to show you!

However, I did have fun tinkering with electronics and simple programming.

What you see in the picture above is an Arduino Uno (top left) connected to my PC via a USB cable and sitting on a base that also holds a breadboard for experimentation. Surrounding it are three “shields” that can be stacked on top of it and controlled by it. Going clockwise, they are an Ardumoto shield for powering electrical motors, a Voice Box shield for synthesising speech, and an EasyVR shield for voice recognition. Also in shot, some tools including a Portasol Technic butane-powered soldering iron, and an SRF02 Ultrasonic range finder.

I’ve tinkered a bit with each of them – enough to be sure I can get them working – but spent most time on getting the Voice Box shield to talk. This required some soldering to attach the ”headers” (the black strips of connectors that allow the shield to be stacked on top of the Arduino) and a couple of screw connectors that speaker wires could be pushed into.

In the process I found that I could still solder stuff, but wasn’t very good. Also that the gas soldering iron wasn’t very good for recovering from bad soldering events, so it’s a good thing I was just a bit untidy and didn’t have any real disasters.

Programming the Arduino is done in a simplified version of C++, based on the Wiring platform, with some of the structure of your program pre-defined and hidden from you to make it easy to get started. It has its own IDE, originally developed for the Processing language. Arduino programs are called sketches, and they control the behaviour of the pins around the edge of the board, which you connect other components to (such as the shields, or standard electronic components).

Personally the first thing I did was look for a plugin that would allow me to work in Visual Studio rather than the Arduino IDE, and luckily there is one. Despite the claim that “serial tools are far better than Arduino”, I did find myself dropping back to the Arduino IDE on occasion for the tool that observes the serial output from the Arduino, as some of the output wasn’t appearing in the Visual Studio version, even though I knew it was there. Also some of the UI idioms it uses weren’t entirely at home in Visual Studio – adding a new Project from the Project menu, rather than the File menu for example. Apart from those quibbles though, the Visual Micro plugin gave me an environment I was happier working in.

The Voice Box shield turned out to be simple to program to. Using standard Arduino library functionality, you create a software serial port that will use two pins to talk directly to the SpeakJet chip on the Voice Box shield; then you send arrays of numerical commands to that serial port. That’s it, as far as technical stuff goes.

Where the work arises is in figuring out what those numerical commands should be. There are some general commands for such things as volume, speed, and pitch, and then more specific commands that represent phonemes that can be used to build up your speech.

There is a helpful program that will do some of this work for you, based on a dictionary of words that other people have already created. If the dictionary doesn’t contain the words you want, however, you’ll have to build them up one phoneme at a time. This applies to perfectly normal words that you’d expect everyone would want an electronic voice to say for them, such as “Exterminate!”, so get ready to spend some time looking up numerical codes for phonemes.

I spent less time with the EasyVR voice recognition shield, so don’t have a lot to say about it at the moment, except that it maintains tables of words it can recognise and the pre-defined words worked perfectly well for me. I haven’t yet investigated customising it, but working within the limitations of this and the Voice Box, I hope it’ll be possible to build up simple dialog trees.

I got the Ardumoto shield to drive the motors on a Dagu Magician chassis. They appear to be a bit weedy, so I don't think they're going to drive that Sevan's Dalek I have lying about, even though it looks like the Magician chassis would fit beneath the skirt, but it’s a start.

So if everything works, why haven’t I achieved more? Well, you may have guessed, but everything working fine separately is not the same as everything working in unison. Although there are quite a lot of pins controlled by the Arduino (6 analogue and 14 digital), inevitably some of the shields want to use the same pins for different purposes. The next step is to do something about this, to which end I’m looking at the Go Between shield from Mayhew Labs; and if in future I want to add many other components so that a robot can interact with the outside world, I will probably end up looking at a Mux Shield.

In future, I might like to play with a Netduino in order to get a nicer programming environment, but I started off with an Arduino because it’s more widely used and supported and I wasn’t going to be faced with the possibility of having to write my own support libraries for shields before I really knew what I was doing.

I’ve had a bit of fun, it’s not been too taxing of my electronics or programming skills, and it looks like more interesting things could be fairly easily achievable.


Mar. 30th, 2012 03:42 pm
ggreig: (Default)

Apparently this has been around for a while, but I only came across it last week: sugru is a silicone rubber putty that comes in little sachets like a largish wad of chewing gum, and overnight it cures into, well, silicone rubber with the strength and flexibility that you might expect. Having seen the interesting examples of its usage on the web site, I thought I’d buy a pack of 12 assorted colours and see what it was like.

The pack that arrived was smaller than I expected – about A5 size – and it was a surprise to discover that sugru had a “best before” date, giving it a shelf-life of about six or seven months. That was a bit disappointing, and although it was prominent on the packaging I hadn’t seen any mention of it on the web site. Black mark!

The other drawback that I noticed fairly quickly, which doesn’t bother me but might be a show-stopper for others, is that sugru is not food-grade – so shouldn’t be used to make anything that will come directly in contact with food. Again, disappointing for a silicone rubber.

However, since I didn’t have any immediate application in mind, I didn’t need a large quantity, and in light of the best-before-date I was doubly glad I hadn’t gone overboard.

I was expecting to have to try to think of a use for it, then when I was working on the Neanderthal head a need suddenly rose. I used some Micro Set to make a water-slide transfer adhere better, but when putting the lid back on to the bottle I burst it! The plastic snapped around the top of the lid, as I tried to tighten it up too much.

It wouldn’t have been my first choice for what to apply an interesting silicone rubber fix to, but it did seem like a good candidate, and timely, so I gave the sugru a go, and applied it as a new grip around the top of the bottle cap, covering the split.

It seems pretty sound, it’s gripped the plastic of the bottle cap well, feels firm to the touch with a very slight give, and is quite grippy. Most importantly, it’s sealed the disastrous split so that none of the precious fluids can escape. For this simple task, it’s done the job well.

It’s not the most thrilling application, but this stuff has a lot of potential, so I thought it was worth a) mentioning, b) sharing my experience and c) pointing out the couple of disappointments I discovered with it, so that people are informed.

ggreig: (Default)

I’ve been on holiday this week, and it’s been a chance to tinker with stuff that I struggle to make time for at the weekends. One of those things is a peg sculpture of a Neanderthal head (pegs à la forensic reconstruction, that is). I found it being remaindered in a toy shop last year when I was looking for a present for my godson.  I figured it was a bit old for him, but something that I would love to play with… (He did get something else, don’t worry!)

I had a choice between this and a gorilla, and according to the advertising material inside there were also a Tyrannosaurus Rex, Julius Caesar and a horse in the range, but I think this is the one I would have chosen anyway – the T-Rex obviously wouldn’t have been 1:1 scale, and a Neanderthal beats old Julius for interest any day.

There were some reasonably detailed instructions inside the kit for reconstructing the Neanderthal’s face, but one vital piece of information was missing – what is this stuff you’ve given me to build the face with, and is it going to set? It was referred to in some places as “clay”, and on the packets as “modelling material”, and it looked a lot like Plasticine.

Without a very definite idea of how the “modelling material” was going to behave, I wanted to have enough slack available to be able to just keep going if time proved to be an issue, so it became top of the list of things to do this week.

Here’s what I started off with; a skull (cream) with some moulded muscle (yellow) and fat (white) on top. Not quite sure why the fat was there, as ultimately it didn’t contribute much to the shape of the face, but I guess I was being informed as well as entertained:

The moulded skull, with muscle and fat attached.

The first thing to do was to cut the red pegs off their sprue and insert them into their matching numbered holes.

Moulded skull with depth pegs now attached

Then the first of the “modelling material” was applied, to bulk up the cheeks. The “modelling material” turned out to behave awfully like Plasticine, as well as looking like it, and I think from now on we’ll assume that that’s what it is. Here the aim was to build the cheeks up until only the small pips on the end of each peg were still visible – the little dots you can see in the picture. I moved to paper towels here as I realised the newsprint was leaving marks on the back of the skull:

Building up depth in the cheek

Next, apply eyes and former for nose, and suffer accusing glare. Eyes and nose were cast in white plastic, with water-slide transfers for iris and pupils:

Nose former and eyes applied

Roll out a sheet of Plasticine to 3mm (roller and depth-graded tray provided) and apply from brow to back of skull:

Skin attached over top of head

Do likewise with a couple more sheets to cover the sides of the skull:

Skin attached to sides of head

Then apply another sheet from the bridge of the nose down to the chin, and form tightly around the mouth and nose:

Neanderthal 021

Apply another sheet from just below each eye down to under the jaw line. This builds up the cheek, and gives it a nice smooth surface, unlike the slightly rough surface built up by hand before:

Skin applied over mouth and nose

Apply eyelids. Ned now looks bored rather than accusatory. This stage was a bit tricky, and the waterslide transfers suffered a bit here, although not enough to be a disaster:

Eyelids applied

Build up the nostrils (compare with previous picture):

Nostrils built up

Form the lips and filtrum (groove beneath the nose):

Mouth built up

Build ears around white plastic formers, remove the place-holder pegs that have been in their place up until now, and stick on head. The ears I made are pretty rubbish and I have a whole newfound respect for anyone who can get ears right, whether drawn or sculpted. The ear doesn’t look too bad in this photo, but I could easily have picked a less flattering angle:

Ears applied

Add final detail to the face; lines around mouth and nose and under the eyes, and dots for pores/bristles. Apparently there was a hair pack for the Neanderthal sold separately, but I couldn’t locate one to buy and decided to go ahead without it. Having found a picture online, I think perhaps I wasn’t missing much:

Final detailing; lines and pores

I lent him my glasses for this picture, to counter Neanderthals’ image of being lacking in intellect. This Neanderthal looks down his nose at me because I neither know nor care what the semiotic thickness of a performed text is.

"Tell me, what do you think of the assertion that the semiotic thickness of a performed text varies according to the redundancy of auxiliary performance codes?" 

So now I have a creepy Neanderthal head to keep about the house and gather dust. Every home should have one! It’s a shame that it does appear to be Plasticine and therefore not as permanent as it might be; so at some point in the future I suspect it will be reduced to its component parts and/or discarded. However, for now, it’s kind of satisfying to have the result of a (very long!) day’s work to look back on.


Mar. 14th, 2012 12:17 pm
ggreig: (Default)

I can identify with quite a lot of this article in which the Grauniad stands up for ’em.

ggreig: (Default)
There's a story on the BBC about a trainee surgeon in Scotland realising it's possible to print models of bones from 3D CT scans relatively cheaply, for inspection before surgery. Still a bit on the pricy side for personal use, but cool to see technology helping to make lives better.
ggreig: (Astronaut)
ggreig: (Ribart's Elephant)

After eating at Pink’s, we went on to Griffith Park, which covers the hills above Hollywood, including the famous sign, and walked up-hill to the Griffith Observatory. The walk is not very long, but climbs at a fair rate, and in the heat of a Californian summer day it was a relief to get to the top.

Griffith Observatory, Los Angeles (Hollywood sign very faint on left)

We didn’t stay at the observatory long enough to do any observing – the telescope is open to the public after dark – but we did sit through a show in the planetarium (enjoyable and informative, although not likely to tell my audience anything new) , then wandered round the museum. The upper floor was mostly devoted to meteorites, and the lower to planets of the solar system. Pluto has not yet been eliminated, although its change of status was noted elsewhere – the gift shop sells “Pluto – Revolve In Peace” T-shirts and baseball caps.

Model of the Moon, Griffith Observatory, Los Angeles

We bumped into a familiar face:

Einstein and me

There was an excellent view over Los Angeles from the top of the hill.

Los Angeles from the Griffith Observatory

When we got back down to the car, a coyote dropped by, and posed for a few seconds on the bank opposite, but moved off just before I managed to press the shutter release on the camera.

Perhaps the coyote was just leaving The Trails, a little Zagat-rated cabin café selling high-quality snacks and drinks for people in the park, including interesting varieties of shortbread (lavender, and fennel and almond) and a variety of vegan-friendly fare including shepherd’s pie, pigs in a blanket, chilli and galettes.

ggreig: (Ribart's Elephant)

The Bell Pettigrew Museum of Natural History, which I visited and reviewed a few years ago, will be open on Saturday and is well worth a visit. Forward notice courtesy of Event: St Andrews, which is collated on a voluntary basis by [livejournal.com profile] flybynightpress (and which has an RSS feed).

ggreig: (Poppies)

It was very still tonight when I got home, with a bright moon, so I thought I’d try some long exposures. The shot of the Plough (and the lights of Angus on the other side of the Tay) was exposed for a minute. The moon over the North Sea was 30 seconds. Both shots are from the same location, looking in different directions.

The Plough


Edit: P.S. I've just noticed that with the minute-long exposure you can see the rotation of the stars in the sky if you click through to open the picture full size, and look closely; the tracks are at different angles, centred around the north.

ggreig: (Steam Coach)

A video snippet from the BBC that I managed to miss earlier in the week, with rather a good look at an electric cab from 1897:


Sep. 14th, 2010 08:38 pm
ggreig: (Blockhead)

I got back to Scotland from a holiday in California visiting [livejournal.com profile] msinvisfem last week. I saw a lot of interesting stuff, so there’s a lot to write about, but based on previous experience – and about 12GB of photos and videos to sort through – it may be a while before it hits the blog here.

Outside the Microsoft Store, Mission Viejo, CAIn the meantime there’s one thing it’s quite easy to write about quickly, so I’ll start with my trip to one of the first Microsoft Stores, at Mission Viejo. There are currently only four physical Microsoft Stores, all in the States, so while the chances of me buying something off the shelf were pretty slim, it was an interesting opportunity to visit and see what there was to see.

The store bears a distinct resemblance to an Apple Store, though a bit more warm and welcoming with varnished wood in place of the sterile lab look. The staff seemed interested and helpful, although being British and just there for a look, I was mostly more keen to dodge them than interact. The hardware was nice to look at, but difficult to arouse much enthusiasm for when I’m not in the market at the moment, either personally or at work, having just got a touchscreen laptop at the start of the summer.

There was one young guy I spent some time chatting to though, who was demoing something I was surprised and pleased to see: Kinect for Xbox 360. If you haven’t hear about it already, it’s due out in a couple of months and it’s a way to interact with the Xbox 360 without a hand-held controller. That’s a big deal for the Xbox, which will help it catch up with other less sedentary game machines such as the Wii, but it’s also a big deal full stop if it’s actually good; bringing sophisticated real-time computer vision into peoples’ homes (also voice control and facial recognition, though those have appeared in home devices such as phones and cameras before). That’s impressive, and – assuming it’s successful – not so much catching up as leap-frogging other consoles.

It’s an impressive technical achievement, but is it really much different in terms of play from a hand-held controller? I’m not really in a position to say definitively, but the difference is that it’s (quote) “full body play” (promotional video). You only need one controller sat in front of the TV screen, and it will track not just the position of your hands or your feet, but can follow facial expressions too. Judging by the promotional video, it can handle two players at once. I don’t know whether more are possible.

I had a quick shot at a ten-pin bowling game:

Ten-pin bowling with Kinect for XBox 360

First of all, you get the machine to recognise you by positioning yourself on a red spot that appears on the “floor” on the TV screen. I had to shuffle backwards slightly to get “myself” on the spot. Once that was done, all I had to do was reach out my arm to the right to pick up a ball; and do what came naturally to bowl it.

I bowled six frames, and had no difficulty picking it up. In fact, in common with my similarly limited experience of bowling with the Wii, it might be a bit too easy; within that six frames I managed to bowl a turkey, which I’ve never heard of before and certainly never achieved in real life. However, it seems there may be room for greater finesse; the demo guy said that once you’ve practiced a bit with it you can apply spin – and all without a hand-held controller!

Six frames of bowling isn’t enough to give a comprehensive overview of Kinect, but it was fun and natural, and I’m quite excited about this development – both as an Xbox peripheral and as a significant achievement for applied computing in the home.

June 2017

45 678910

Most Popular Tags


RSS Atom

Style Credit

Page generated Jun. 25th, 2017 12:09 am
Powered by Dreamwidth Studios