Tonight’s  speakers were Tobie Kerridge a researcher at Goldsmiths who talked about a projects called Material Beliefs, Yuri Suzuki a product designer and musician and Dominic Harris from Cinimod Studios who talked about designing and building a UFO - great project.

I was very lucky to get hold of a ticket for the event as they go like hot cakes. Before it started I got chance to have a quick chat with the guys and they told me they are looking for a more permanant venue so they can meet the demand.

Thanks to Liquid for letting me take along the studio camera and help Chris, Joel and Andreas document the event. Joel is planning to edit all the footage together and put it live on their website shortly so watch that space. For more information on the event visit their website

The journey began in Whitehaven. It was already raining. After a few disagreements with traffic and not too cheerful residents (they shouted things too rude for this) we headed across a nice flat disused railway. The initial pace was fast, with frequent rests to keep the group together. The first hills were a challenge but once my legs realised what was going on, I found myself getting up them in good time. The initial hill was rewarded with a long downhill where my friend Dan clocked 37.7mph.

As the rain grew heavier we reached Whinlatter pass. A stupidly tall and steep hill, the first serious challenge of the journey. After a brief accident with my chain I concentrated, looked at the floor and kept turning my legs until i reached the top. The feeling of success was incredible. If it wasn’t so windy and cold, the view would be amazing too. We reached the top of a woodland section with 18 miles to go till our Bnb. It was already 2pm. If we were going to make it in time then we’d have to pick up the pace… which we couldn’t. Uh oh.

After a fantastic, insanely fast off road section through a forest, we arrived in Keswick for 4pm. Our bnb was expecting us but we still had 15 miles to go. Panic set in amongst the group and this last section of the day was a grim push to get there before we lost light. Arriving in Penrith I promptly got lost on my own. When we finally reached the Bnb the host pointed out several of us had the early stages of Hypothermia. Awesome.

Considering the awful end to the previous day, we set off dissapointingly late. This would have been fine if the Bnb host hadn’t just lent me a jacket because he “didn’t want people staying at his place and then dying on a mountain”. The radio was reporting 60 severe flood warnings, one death and “the worst flooding to hit the north east in years.” This is when it hit me. We were about to cycle into something that could put us all in geniune serious danger… oh well! We can still try our best!

Upon reaching the base of Hartside Pass (the big scary 1900ft one) disaster set in. The winds were already stronger than anything i’ve ever seen and roads around us were filling with water. It was actually like the world was falling apart around us. Eventually the decision was made to reluctantly call things off. Under the circumstances and abilities of the group, to try and scale the mountain would be like organised suicide. Everybody was heart broken. I’d been training for months so couldn’t accept what was happening. I’d found out 3 others had made it to the top so guess who decides to go it alone. Muggins.  The gusts were reaching 60mph and the rain felt like a sand blaster. The cold was unbearable and I couldn’t hold my bike in a straight line. Possibly the most stupid, dangerous decision in my life but worth every second of it!

There were 4 of us left from the original 12. The others were collected 2 hours later in emergency blankets from a farm somewhere near Renwick. We had to decide what to do but we knew being on the top ofa mountain when the cafe shut was a bad idea so all we could do was push on. Much as Hartside was built up to be the biggy, we were still in the Pennines and it had plently more peaks saved up for us. Garrigll was particularly delghtful as the entirity was spent inside the cloud layer so we could actually tell where the top was. This place was so barren and disastrously windy, it was like being on mars. Shortly after this we had to stop, simply for our own safety. A rescue mission was in progress and 2 scout leaders came to collect us in a landrover from a pub in Nenthead where we were escorted to the Bnb. On the way we drove past houses underwater and trees thrown across the road. We’d just done the hardest part fo the C2C in the worst weather ON RECORD.

As we had to be rescued, we were taken back to Nenthead to resume the journey. This meant we started the day with 3 more peaks to cover and a total mileage of 60 mile or so. By now I’d seriously damaged each of my knees so a feeling of apprehension was shared by all. I just grit my teeth and screamed a lot. Somehow we covered this final peaks (including the higest) in just an hour. The knowledge that we had a support vehicle made us unstoppable, having prearranged rest points to stock up on food and painkillers. At one point I was fed painkillers out the landrover window while moving. Ibruprofen, Paracetemol and sheer stubborness were the only things keeping me going! Theres much less to say about the second half of the day as things got awkwardly flat. We averaged 12-15mph as we rocketed in a straight line towards the coast. By now the pain was unbearable but there was no way of stopping me after what we’d been through. Newcastle was possibly the hardest pasrt of the journey as the end was in sight, but Newcastle was 12 miles long! Debris was everywhere from the previous days floods that had luckily dropped to let us pass (we did have to bike through a few 2ft lakes) Before even realising where we were, the support guys were stood on a hill taking photos and that was it! We reached a signpost announcing we’d done it so I threw my bike in the sea!

19 Peaks later, 149 miles, 60mph gusts and over 48 hours of constant torrential rain. 8 people less and two annihilated knees… wow

A massive thanks to all who sponsored me and an extra special thanks to all of those who helped me get to the end. The other jackrabbitt riders, the fantastic three and especially the support team. Without them i’d probably still be on a mountain somewhere.

Inspired by his managers previous fundraising wetsuit exploits, our newest designer Tom will be tackling the coast to coast bike ride this September.

He will be cycling the 140 mile route from the Irish Sea to the North Sea over 3 days to raise money for the MS Society, helping 85,000 people in the UK who are affected by Multiple Sclerosis, and Wheels4Life. A non-profit organisation providing bicycles for people in developing countries.

Training has been going well for the last 2 months and he is confident he will return to us in one piece! Just about!

Please support the MS Society and Wheels4Life and visit Tom’s donation pages

Use the green box to have a search round the finished piece and see what goodies you can find. Enjoy

Following on from my blog about Nautical Crosses and using simple game AI to make decisions, I thought I’d go a few steps further and include an different game that not only uses much more advanced decision-making techniques, but also uses code-driven animation to simulate a traditional board game.

The game is based on one I can remember having on my Nokia phone many years ago called Bantumi. At a time when we were churning out idea for London-themed games this one called ‘Londinium’ came about – it’s Britons v Romans in a traditional-looking board game set up. You take the role of the Britons (unless you choose to play a friend in 2-player mode) and you choose a piece to move anti-clockwise round the board with the aim of landing on and capturing your Roman enemy. The piece can move by the number of points indicated in the square where it’s standing, and as you move round one point is left on each square you touch. Capture an enemy character and you win the points from that square. I suggest you try it and work it out for yourself – it’s not as difficult to play as I make it sound, and it can quickly become addictive.

The software uses weighting to decide the Romans’ movements (as with Nautical Crosses) and there’s a definite aggression about the Roman tactics. If a Roman can safely take a Briton’s piece then it will, otherwise it’ll look for either a defensive move or will look a few moves ahead to see it can trap a Briton – bear in mind that if one of your pieces is blocking a final landing square ahead then you won’t be able to move out of the way and will face capture.  You can win back captured pieces for every 10 points you achieve.

This game was never finished to a commercial level so there may be a few ‘rough’ edges. A former colleague of mine, Dan Dawson, created the graphics for the game. There’s a lot more than decision making on show here of course, including using code to animate objects as well as scaling objects in a pseudo-3D world. For those familiar with Director, I should point out that all the game play happens in one frame on the score and is included in a single frame script, other than some simple animation code to scroll through text which sits in behaviours on text sprites.Originally the game was made for use with a keyboard and used arrow keys and ENTER to scroll round the board. I have, however, just updated the code to accept mouse clicks; so if you prefer you can just click on the piece that you want to move.

For commercial reasons, the source code isn’t available but I hope playing the game demonstrates the principles enough to make this article worthwhile.

I recently read somewhere that winning the UK national lottery is easy. All you have to do is spend £1 on a ticket with the next draw’s numbers on it. I also read that the odds of managing to do that are calculated thus:49/ (6 x (49-6)) or 1 in 13,983,816 to be more precise.

Having treated both of those facts with the contempt that they deserve, I slowly started to realise that my tactic of spending a quid every so often on 6 numbers that match my birth day, birth month, house I used to live in, house I currently live in, my wife’s birthday and my favourite number probably isn’t the best way to hit the jackpot. Anyway, to cut a long story short(er), the focus of this lab entry is to let you see if your lucky numbers would ever have won the big one, dating back from when the UK lottery started in November 1994 right up to present day.

Why? To demonstrate how easily shockwave allows us to create a web application that will draw on data someone, elsewhere on the Internet, maintains for a completely different purpose.

You’ll most likely get a shockwave security alert. I promise that it’s safe to allow permissions, but if you don’t feel sure then not allowing it will merely restrict the available results to those available when I wrote the software.

There’s an interesting article on avoiding this shockwave security alert for developers at http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_12777 At the time of writing, there have been 1277 lottery draws. This shockwave movie will do its best to update all winning draw numbers before offering you a verdict on yours. However, should the internet throw a huffy fit and it can’t find the results, it will at least be accurate up to the draw on Weds 19^th March 2008.

Pick your 6 balls and see if you’d have won. Click on a selected ball to deselect it and then alter your combination. I offer no consolation prizes if you find out that the week you forgot to put them on, your numbers came up, other than the advice not to think about it too much!

If you thought winning the real lottery is hard, then picking 6 numbers that would have won sometime in the past isn’t easy either! In fact, I suppose the odds will be 13,983,816 divided by the number of draws, so as I write this there’s a 1 in 10,950 chance. So, good luck with that one!

Thanks by the way to Richard – his website http://lottery.merseyworld.com/ is a font of all knowledge for lottery related matters, and is indeed the source I’ve used for the winning lottery numbers.

And thanks to our new junior designer, Tom, for sprucing up my attempt at graphics in the shockwave movie.

If you want to test the software with some sets of numbers that have won previously, here’s a small selection:

02,03,12,20,33,49 - 03,09,10,12,46,47 - 04,18,22,27,44,48 - 07,10,23,28,30,48 - 08,14,24,28,39,45 - 11,16,20,22,27,46

Right… I’ve opened 3Ds Max for the first time ever and I hold a degree in 3D modelling and texturing… so this should be easy right?

Wrong.

I’m trained in Softimage XSI and I’ve just opened up a completely different looking piece of software with an entirely new layout.
I may as well be learning to knit.

This will be the first of several tutorials, following the modelling of a car from start to finish. As I am new to this software myself,
I will be learning as I write these so lets see what I can teach us!

In this tutorial we will go through the basics of the interface and how to set up your blueprint images. Its always useful to have reference images
behind your model to keep proportions and shape in check.

Before diving into the 3D bit, find a blueprint of a simple car online and use photoshop to cut it into seperate images. www.the-blueprints.com has over 23,000 to choose from.
I use guides, snapping and copying each view into a new file.
Save them as something obvious like “side.jpg” “front.jpg” “back.jpg” and “top.jpg”.

Now for the 3D.

First things first. Learning to get around the interface, which on first appearances can be quite daunting.
The UI or User Interface comprises of a standard windows menu (orange), central viewports, animation timeline (red), a small navigation box (green),
a main Toolbar (yellow) and a Create/Modify Panel (blue)

As we are only modelling, we can ignore the timeline and windows menu and we’ll start with viewport navigation using the tools in the green box.

Put simply, the central viewports display our work from a variety of angles. This allows us to make changes from one direction and see the
effect it has on others. Left clicking in each window will make them “active” allowing us to make changes in them and each angle can be changed
to another with a right-click to the top left corner and selecting View.
Dragging the borders between each of the viewports alters their proportions and Alt+w maximizes the active viewport to full size.

The following icons can be used to manoeuvre around the scene, all of which have a mouse equivalent on the middle wheel/button.

- Scroll wheel - zoom

- Middle click - pan
- Alt + Middle click - rotate

(If rotating in one of the flat views, a strange three dimensional view appears which has NO perspective. This and all viewport movements can be
undone with Shift+Z)
Right Click - Deselect tools

We will set up the blueprints as textures on flat planes so lets begin to make those. The creating/modifying and generally doing anything useful
menu is on the right (highlighted blue).
The tabs along the top represent from left to right: Creating, Modifying, object Hierarchy, Motion, Display and Settings. For now we’ll just need
the creation and modifying tabs.

Within the Creating tab, the row of icons below is the type of object that we wish to create.

Under primitive objects, select a plane and left-click-drag a shape into the perspective viewport. Any size will do as we will modify this shortly.

Now that we have an object in our display, we need to get to grips with how to move, rotate and scale it.
All of which can be found on the top Toolbar (highlighted yellow). The first buttons we need to know are as follows.

- Undo/Redo (Ctrl+Z Ctrl+Y)
- Select (Q) Select by name
- Selection draw method
- toggle selection via intersect or surround
- Move item (w)
- Rotate item (E)
- Scale item (R)

(in all cases click on different coloured axis of the gizmo to manipulate in specific dimensions. Blue for Z, Green for Y and Red for X.
Click between axis to manipulate in two or in the centre for all three)

IF ANY GIZMOS ABOVE FAIL TO SHOW, PRESS X.

Our first plane is sitting correctly but we need one for the side and front.

With the first plane selected and in move mode (W),
hold down shift and move the plane. Upon letting go, a box will appear. Leave the settings unless you wish to change the name and hit OK.
You will now have an unlinked copy of the first plane.
By using the rotate tool and a third plane, align the the two new planes to their correct angles. For more accurate 90 degree rotates you can use the
information boxes at the bottom of the screen.

Now, in explorer, find your image files and get up their properties so you can see their dimensions. Back in XSI under modify>parameters input
these dimensions to rescale each plane to its correct size, ready for the images.

Creating textures in 3DS Max is comparatively simple compared to XSI. Press on the Toolbar or (M) to bring up the materials menu.
Each sphere represents a texture so simply find your images and click and drag each one onto a seperate sphere. This will make 4 different textures
for your four different car views.

To apply these textures, simply drag them from the material menu on to the plane of your choice. As we have already
made the planes the correct size, no tiling should occur.

Now all we need to finish off is align each plane with the others. Select one object and press . With this align tool, select the plane you are
aligning with and a menu will appear. Basically click buttons until the planes have aligned correctly. I’ve yet to make head or tail of which axis make sense
so its very much trial and error.

Another good idea is to select to bring up the layer panel. With all the planes selected with Ctrl, create a new layer and move selected onto it.
Checking “freeze” will make them unselectable while we model. Very handy indeed.

Hopefully atleast some of that made some sense because next time, I’ll be trying to explain polygon modelling!

vimeo video link

We created this experimental Music Visualiser in processing & minim sound library.

Path finding is the art, or is it theory, of calculating the best route from A to B. This isn’t always straightforward, particularly when solid obstacles are placed in the way.

The A* algorithm is one of the most commonly used methods in games these days, and to avoid this web page turning into a book, I’ll keep this overview of the techniques involved as simplistic as possible - in doing so, I may take a few liberties with some of the more in-depth stuff so please excuse me if I miss anything out.

The little demo game above shows my experiment in using A*. To test it, you simply drag the green target to a new location and click the ‘GO TO TARGET >’ button. The little yellow dot will then speed off, making its way to the target.

How does this work? Well, for a start the dot sees the maze as a grid made up of squares - in this case each square is 5 x 5 pixels. A higher resolution (e.g. 1 x 1) would have led to greater accuracy, and 8 x 8 might have satisfied the true tile-game traditionalists, but 5 x 5 it is. The lack of resolution adds to the retro feel, and also helps to show that there is actually a grid involved here, giving slightly clunky turns around corners.

The grid is referenced, as in all tile-games, as a list (or array). Within the list are sub-lists. This grid is 128 squares wide x 96 squares high (640 x 480 pixels divided into 5 px segments). The list has 96 items (one for each row), and each item is in turn a list of 128 items (one for each unit in the row). Examining item [10][128] for instance would reveal the value stored in the last item of the 10th sub list. The values stored in the list are either 10 or -1, representing clear space (black) or solid walls (red) respectively.

When you click ‘GO’ the little dot starts to calculate the best way to the target. In simple terms it will start from it’s own location, and move out a square at a time, either accepting them as possible moves, or rejecting them either because they’re a solid wall or they ‘cost’ too much. It calculates the expense of a square by adding 10 points for each square it would travel from the starting point to get there, and also adding 10 for each additional square it ‘thinks’ there will be left for it to reach the target point following the current path (this is essentially a well-educated guess - google heuristic’ and you will find plenty of people who can explain it better than I can). It will do this until it reaches the target square. By this point it’ll have several likely routes to take, so it does a quick track backwards with the same calculation and arrives at the start point having decided the ‘cheapest’ route. All of this without moving at all! Once a list of squares to visit on the journey is ready, the dot simply travels through them.

You might wonder why I bothered giving clear squares a value of ten, when one would have done just as well. The reason is to allow scope to introduce other types of squares, which may have a lower cost when used on the journey, paths for instance as opposed to rough terrain would have a lower cost and therefore a lower value assigned to them. It’s using the algorithm in this slightly more advanced way that makes games more interesting and, when workload allows, I’ll work up a more game-like example of path finding across different terrains, as well as tidy up my code a little to avoid some of the slight errors of judgement that the dot makes occasionally.