Ever since the first LEGO Star Wars game came out, I’ve been trying to make a 3D recreation of a LEGO minifig that has the same range of movement – bendable legs, arms, hands, that sort of thing. Then I found this custom LEGO model of a Dalek by Flickr user kaptainkobold. I thought “why not?” and with the help of the LDraw library (open source virtual LEGO pieces) made a virtual version.

The result was a homage to the Doctor wailing on a Dalek with a wrench in a recent episode:

The Daleks aren’t having much luck with serving tea, are they?

Artefact One

My research topic is “how do you convey a sense of believability and narrative in virtual environments?” Now admittedly this is an odd area to perform research on. What exactly do you investigate?

Just over a week ago, however, it occurred to me that how an environment is set out is rather important in making it believable – if you’re not inclined to stay and look around a place, then any effort put into the buildings and other elements goes to waste. The originator for this thought came from an old “God game” called Populous: The Beginning that my brother used to play the demo of and the way that, particularly in the first level, he would set out the houses in a neat grid pattern. This always bothered me for some reason and when thinking over it for this artefact it made me think back to my visit to Coventry last year, which I didn’t enjoy at all. And over the course of the week, I realised why.

Coventry was extensively re-developed after the Second World War, and one of the results was that the main shopping part of the town centre is laid out something like this:

It’s all straight lines and tall buildings with few landmarks to tell you where, exactly, you are. And as a result I got rather horribly lost when I was there. There was also something about it that I just didn’t like – I’ve always loved going to cities like York and Durham and Lincoln because of how much there is to explore – in two trips to York, I’ve lost count of the number of side-streets I walked down, and I still haven’t covered more than half the city centre.

And then I remembered what it was like going through Ravenshead when I was young – it actually took me years to realise that there were side-roads going off the main road and I actually thought that the village only consisted of that road, because we always drove straight along it.

So with this in mind and going back to my old Geography lessons, I located some free building models (found on this site) and using a program called Role-Playing City Generator (which generates village, town and city layouts), 3DS Max and Adobe Director, I constructed examples of a linear settlement and a nuclear settlement (not sure of the correct term) that could be navigated in Shockwave and put them as a download on a page of this journal with questions to gauge people’s reactions to them. Then a friend kindly offered to host them on his server so that people didn’t have to download it (click on the images to load the simulations).

From the top, the two “villages” look like this:

Some of the responses I got I was expecting, particularly those saying that they spent longer going around the second village. With the first village, as one person pointed out, you’re likely to just walk straight through it and possibly not notice the side-road, as is what happened with me and Ravenshead. It occurred to me that linear settlements are the environmental equivalent of the long corridor in games – you just want to get through it as quickly as possibly. It also reminded me of a chapter in The Art of Game Design by Jesse Schell where it talked about if you placed a line on the ground going straight ahead or to something, people would go straight along it.

Several responses were gained whilst in an IRC chatroom. Whilst this didn’t garner many individual results, it did raise a couple of points that I hadn’t previously considered. There was a consensus that the second village was more interesting “because it has these twists and turns in it” (Bert Loos, although he also admitted to nearly getting lost in it), and a couple of others said that they preferred it because they were adventurous people. One person, Aaron Bulger, gave the most thorough response:

I immediately began by following some of the paths in the hills, then returned to the village, in which my navigation was more of a meandering. I spent much more time off the beaten path in the second and was interested to see how the paths fit together.

The Art of Game Design addressed this as well; by adding more lines on the ground, people were more likely to veer off in different directions.

However, there was some difference of opinion as to the reason why Village Two was preferable, and it unexpectedly fed into the narrative part of my research topic.

One person, who went by the username “Dog”, said that the second village looked “more friendly”, and gave a rational for why this was: “Order+uniform=bad. Zigzags+non-uniform=good.” Another person, who went by the name of “Blub”, preferred Village Two for a very different reason: “[it] gives you more possibilities to escape,” whereas with the linear one “it seems to be that you can only escape in two directions.” This was because they were a really paranoid person, and it wasn’t something I hadn’t even thought about. If you were in a linear settlement there isn’t many options for you to hide yourself if you needed to, but in a place like Village Two you have far more corners to duck around and lose someone. But the reverse could also be true – someone could be hiding in the second village and ambush you, and in the first one they would have little opportunity.

The results of this artefact were far better than I expected. Without any prompting from me, some people inferred some kind of narrative on the environments, and their reasoning for their choices was very insightful. To build on this unexpected outcome, my next artefact will quite possibly look at how the sizes of buildings can induce a sense of foreboding or welcoming, such as with Coventry, being surrounded by tall buildings on all sides was somewhat claustrophobic.

Final Visualisation Project update

Sorry sorry, this slipped my mind. Didn’t help that our internet went a bit doo-lally for a few days (a fault at the exchange which, despite affecting the whole area, no-one reported for at least six hours until we phoned up).

Before I go any further – I have gone through the camera match tutorial, however my copy of Max was being slightly awkward and making adjusting proved nigh impossible. However, I do understand the basics as I can recreate things from photos quite easily, and will do so if I’m a bit stuck for something to do.

So, my visualisation project was recreating the palace of Xerxes from the Fullmetal Alchemist manga. Which went… Well, not disastrously wrong, just not as good as I wanted it to turn out.

As you can see in the last five, the shadows were being fickle again. And the interior dome… *cringes* I have no idea what went wrong, but Max would not let me fix it. Every time I tried: “Max has encountered an error”. And this was like 5 times in a row. I wasn’t about to push my luck really in case my laptop started smoking like a pop tart.

I think I probably forgot to alter the colour on the pillars, and lower the tile size of some textures, because it looks really nasty when viewed close up. And I mean pixellated nasty.

But overall I am rather proud of it – roll on the Xerxes flashback so I can compare!

Final Animation post of the year.

Okay I admit it. This should have been up before now. In my defence I hit a sort of burnout after hand in and combined with me being scatterbrained it ended up low on the list.

I’ll say this: modelling and rigging characters = piece of cake. Wiring parameters also equals piece of cake. Cameras and x-referencing scenes to avoid a program crash (I hit double figures during this. You do NOT want to know how visualisation went in that regard) equals a pain up the arse. I also learnt to next time not model anything in excess of 100 units tall because this makes the sets you need to build ridiculously huge. Once you’re hitting 4 digits in the XYZ coordinates it starts getting unbelievably silly.

The best thing I ever learnt during this project was just how damn useful hiding stuff is. Even if you’re reaching stupid amounts of polygons, hiding stuff you don’t need is a lot less taxing on the processor. The best thing I learnt though on top of that was that you can hide some of the polygons of an object. It was an absolute godsend when using morph targets let me tell you.

Anyway, here are screengrabs, in order, of the three featured characters’ rigs, one character that never made it into the final animation (the proportions are considerably off on that one), a couple of the two ‘sets’ and finally stills from the animation itself, where it becomes obvious that the lighting went bonkers.

The final animation has no sound effects (except for a couple of “voices”), instead I used royalty-free music from this site. I think I’ve also “Gainaxed” the ending (meaning that it doesn’t really make sense – “Gainaxed” refers to Studio Gainax’s ultra-confusing ending to Neon Genesis Evangelion). Also, I swear the shadows were turned on for the lights. *checks* Yep, they were on. They just refused to work how I would expect them to work (like, you know, actually cast a shadow on the other objects!). Also, can I just say that rigging and animation a 4-legged creature is very tricky? I may know a lot about animal biology, but trying to get it to walk right was still a serious effort. Anyway…

After The End is set in a world where humans are long gone, leaving being a myriad of weird creatures, which are constantly menaced by these robotic “angels” (I never figured a name). In this, it all starts when a strange wraith-like creature called Dormin happens across a broken piece of horn…

Couple of fun facts – 1) The reversed dialogue I used for the Angel is part of the Zagreus poem from the Doctor Who audio of the same name (yes, I am a total geek) and 2) Dormin’s arm does a rather bizarre 360 at one point (the bit when he reaches his hand out in the tunnel). It’s hard to spot due to the lighting, but it’s the only time it does that in the whole thing (I didn’t know how to get the elbow to work with a look-at constraint).

Animation Update

In which I start reaching new levels in downright weird.

Fair bit of the Easter holiday was spent working on my other main character, an “Angel.” The reason it took so long to do it was down to several glitches cropping up that required me to go back to a previous file (good thing I’ve gotten into the habit of using the Save As Copy shortcut), the reaction manager causing crashes (about 10 in the space of 15 minutes, that must be some kind of record) and the problems of not being able to remove any constraints you’ve added except for Undo. As it stands, a character that’s mechanical and doesn’t need bones is even more awkward to rig than one does.

Will add commentary later.

Games Project.

Whoops! Forgot that I hadn’t posted anything to do with this.

So, I decided my level would be a Steampunk airship hanger, because well, it was an excuse to draw airships in seminar. Anyway, I did some sketches of a layout and some design elements.

shipsketches sketches

Yeah, excuse the confusing level map, I haven’t found that book of tracing paper I got with an animation kit a few years back yet. Rest assured I have a rather good idea of how it will look up here. *taps head* Just the dimensions that I have trouble with.

I went fishing for airships and steampunk on the net – Final Fantasy is a goldmine for this sort of thing. I admit I may have been channelling Final Fantasy IX with the airship sketches as I love that game. I also found some awesome pictures of industrial piping and stuff from factories.


I’ve been building the airship in 3DS Max as hopefully the right size to put into the level. However, I’ve quickly discovered that getting it into Unreal Editor isn’t as easy as you’d think.


Here’s the rough airship, looking very… ship-like. The little dark blob in the bottom left corner is my character height guide. Now, when I put it in the editor, I got this:


Looks odd, but I only wanted it for judging my placement (I’ve already mucked up the floor, the pit’s not meant to be all the way along), I can add the final thing later. So I run the level to check the size in-game and this happens:


Holy smeg I’ve got a ghost ship! I have no idea why it’s doing that, and I’ve been looking online to try and find the reason (the book only covers Maya, and while I may be good at translating from one program to the next, I have limits). May have to rethink being able to go inside the ship (but I’m sure as hell making sure you can go on the deck) but once I’ve figured it out I’m sure it’ll be easier to do. I’m currently trying out unwrapping the UVW and giving it a proper texture file to work from. Slightly fiddly, and I haven’t even got to the mast yet.


I’m still incredibly amused at the fact one of the set of faces turned out like an axe. I’m sure I’ve seen one of those in World of Warcraft.

Erm… other stuff when I can sort through my files (I have a fully rigged character for animation done, pending a couple of tweaks).

Character Project

Just to show I have actually been doing something, here’s a progress report for my animation project. Visualisation will have to wait till I can get the stills from Ed (and after we’ve done this blummin’ presentation – am I the only who finds having one group doing Victorian architecture and another doing Industrial Revolution architecture (my group) to be a bit silly and redundant seeing as they’re around the same time period?).

I started off with this:


Didn’t get very far with modeling it as you can see, because it felt rather… safe? Uninspired? Anyway, I’ve filed it away for future use – I’m particularly fond of the rough sketch in the bottom left. I also quite liked the quick (and unfinished) speed paint I did:


But after watching Jojo in The Stars in our animation lecture last week, my brain took a decidedly weird turn, and I ended up creating this guy, who is now almost fully modelled.

creaturesketches creaturetopcreatureside


Things got odder when I found this little flash app called SCRIBBLER – which takes a drawing that you create in it and makes it scribbley (is that a word?).





I ended up playing around with it when I really should have gone to bed, and came up with these… things.


It’s like walking the pencil on steroids. And when I started to use some as a basis for extra designs, things got decidedly weird.


I’m starting to think my subconscious goal for this project is to see how much I can creep everyone out.


And this was me doodling when I should have been listening to the lecture yesterday afternoon. Although apparently they’ve proven that doodling actually helps you concentrate (not sure if it still works like that when your mind starts thinking about things like plot and visual style mind you). In case you’re wondering, they’re sized relatively except for the one on the far right.

Memo to self: Look at some other themes because this one is a bit on the narrow side.