Wednesday, 17 December 2014

Character modelling sheets

Continuing with my character development I sketched some modelling sheets to help when translating my designs to 3D. The game will have a number of ranks per character class and this will be reflected in the way a character looks. A high ranking 'Gentleman' character will be wearing the most decorative/ornate clothing of the period, whereas the 'Bandit' will be start poorly dressed and become more dishevelled/menacing as he levels up.

Character modelling sheets for different character ranks.

Thursday, 11 December 2014

Character portraits and silhouettes

Following on from my research I began working on some character designs. In these early stages I wanted to get a feel for how my characters might look and act. The Gentleman will be refined and well trained, whereas the Bandit will be dishevelled and more of a bruiser.

Character silhouettes and concept sketches.

Using silhouettes is a great way to quickly capture character ideas during the initial stages of a design. The silhouette can also inform the viewer of the attributes of a character, especially important in a fast paced FPS. It is also a technique used in character portraiture of the period, as such I want to incorporate into my game as player portraits, and other UI elements.

A mockup of a loading screen for my game.

Thursday, 6 November 2014

Game design mood boards

To keep my design idea in keeping with the period in which the game is set, I carried out a lot of research. I collected imagery relating to the two character classes (the Bandit, and the Gentleman), the planned environments, and the associated props/assets for both. Some of this research was collated online, whilst some of the environment imagery was taken on location at a local National Trust site. I've brought together a selection of these images to create mood boards that will help me keep my designs consistent as I develop the project.

Character and environment mood boards.

Friday, 25 April 2014

Pirate Game Competition: Live Brief

Part of our brief was to find a competition and work on it as a live brief. At I found an old competition that suited my skillset and gave me the chance to continue my UDK training. 
The challenge was to create a visually appealing and interesting environment using UDK with the custom assets created by WarrenM.

I started by collecting reference imagery based around the pirate theme. This consisted of photography from real pirate coves, concept art and scenes from films such as Hook and The Goonies, and images from the granddaddy of pirate games The Secret of Monkey Island. The images I found were often set at twilight with foggy dark blue hues, lit only by the lights shinning from the windows of the tumbledown wooden shacks. Other common elements were rickety gangways, galleons and crows nests all of which seemed cobbled together from ill-fitting wooden planks. 

Starting with the default night-time lighting in UDK, I dropped all of the assets into the scene and began 'kit bashing' them together to explore the different assets I could create from the limited selection we were supplied with.
The assets leant themselves well to objects like boats, chests and walkways all of which would fit with the theme of the challenge. Of course with the wooden plank asset almost any structure could be built, although this would mean building them plank by plank, but in doing so would give that tumbledown feel.

I was inspired by the spiral shape and texture of the rock asset, it reminded me of fossilised coral which leant itself well to a sea cave. By scaling, rotating and merging the rock asset I decided I wanted to make a secret pirate cove with branching walkways and splintered shipwrecks around the entrance. The rocks I've started to clump together have begun to look like a rock monster so I could take this in a completely different direction to my initial idea, whilst remembering to stick to the theme.

I worked on a few more structures to create walls and archways, with only one rock asset it was tricky to reuse it without the repetition being obvious so I tried to make the spiral forms flow into one another to look like folds in the rock. If any of the assets were scaled too large you could easily begin to see the pixelation of the texture. I placed the rocks to rough out the initial outline of the cove to estimate the scale and height.
By duplicating, scaling, and rotating part of the arch I saved time creating the domed shape of the cove. It doesn't look great from the outside but it'll only viewable from the interior anyway. If I was building a structure like this for a game I would use a single mesh to keep the poly count as low as possible, but that was against the rules of the competition, and as we were asked to create concept pieces polygon count wasn't an issue.

I created some terrain for the floor of the cove using the landscape edit tool in UDK and then using the brush tool to create a simple slope towards the back of the cave. The interior was staring to come together, I used some of the rock structures to create stalactites and stalagmites to give the cave a more natural look. I removed some of the rocks in the side of the cove to allow some of the moonlight in and add some visual interest to the scene, I'd like to experiment with the volumetric properties of the light to strengthen the light shafts that are already there.

With the environment roughed out I started to create houses for the interior, I imagined this cove to be a colony of marooned pirates that have built all the structures from pieces of shipwrecks that have washed into the bay. I tried to make the structures as modular as I could to same time by reusing assets I'd already created to make new ones, using the grouping functions in UDK to create groups within groups really helped to speed up this process.
I made a few more buildings adding detailing such as the ships wheel to suggest that cobbled together feel. I also wanted the environment to have multiple levels so added balconies to one building that I wanted to string together with chain link gangways. Looking back on my progress I really should have added some randomness in the width of the planks, like I did with the length of those on the roof to get a more realistic feel. I haven't worked on the interior of the buildings, but I'll probably do one as proof of concept. I'd also like to have more randomness in the window shapes of the buildings, a round window would look good as if it's a salvaged porthole.

I began to place the assets around the edge of the cave as if they were built up against the wall, whist paying attention to randomising the assets by mirroring and rotating them so the repetition was less noticable. I grouped a point light with the lantern asset supplied to place in each building, this simulates the same lighting that I'd seen in the pirate images from my research.

One of the larger structures I worked on could function as a tavern or meeting hall, acting as the focal point or hero object in the scene. I created additional details, such as the crows nest and ladder by manipulating the plank asset, to enhance the theme and salvaged look to the structure. The only BSP brushes I've used in the scene were for the mast, keeping within the rules of the competition. However I had to use the plank texture provided which was noticeably pixelated when applied, it won't matter as long as I keep the asset at a distance from the camera when I take my final shots, but this would be an issue in a game level.

I placed additional environmental details such as the leaves around the gap in the cove where light would be available for plant growth. As well as adding more visual interest I hope it acts as a signpost to the viewer that there may be an environment outside the cove such as a tropical jungle.

I used the plane and textures we were given to create the sea for the environment. I placed more assets to suggest numerous shipwrecks such as the decaying ribs of a galleon, small boats shattered against rocks, and submerged masts. 
I continued to work on the scene thinking about the people that may inhabit it, how they move about, and the structures they may have built. There needed to be more verticality in the environment so I created a chainlink bridge asset that I could use to link buildings of different levels. I was keen to have walkways around the stalactites, in a game this would allow for multiple pathways in the level making it more interesting, and allow the player to get different viewpoints of the environment. Placing lights at the end of some of the walkways highlights them to the player to let them know it's an area they can access.

The level was starting to come together, however due to time constraints I was unable to put the finishing touches to the environment. I underestimated the time it took to create the assets, mainly as they were built plank by plank. The scene was also becoming very dense, at this point it was taking half an hour to generate the light maps to be able to preview the lighting in the scene. As proof of concept what I had fit the themes and requirements of the brief, and the beauty shots below communicate the idea I wanted to convey, they just need some extra work with detailing and lighting.


Sunday, 20 April 2014

Game Character Update: Earthworm Jim

As part of my 'Alternative Techniques' module for college I was tasked with creating an updated character sculpt of a retro game character. It was a good opportunity to continue with my Zbrush practice. It was tough to find a retro character that hadn't already had some sort of fan update, so I decided do go with one of my favourite characters, Earthworm Jim.
At present I've only done a simple 2D to 3D translation of the character. It's still work in progress as I get to grips with Zbrush.

 I'm going to need to retopologise this mesh as I've had some difficulty with the topology around the arms, but he's coming together. As for the modern day take on the character I want to suggest a Crysis-esque Nanosuit.

Once I've refined the anatomy and topology I'll be able to define the muscle fibres of the suit and add a carbon-fibre material through the texture work to achieve the look I want. With this being a mini-brief it's something I'll have to come back to but I'm happy with the progress I've made so far, especially the opportunity it gave me to refresh my Zbrush skills.

Saturday, 19 April 2014

Weekly Targets

  • Research retro game characters with the intention of creating a modern sculpt using Zbrush
  • Find a competition/live brief to work on as part of my Work Based Learning module
  • Refine/update my online portfolio

Friday, 18 April 2014

Lego Me!

I've been trying to come up with a few different concepts for my logo design. I really wanted to make a Lego version of me, if not for the logo then just a bit of fun! Here's the finished article...

I experimented with a few different concepts using the beard texture I created for the model.

How about the Warhol? Too much?? Definitely!

I also like the pixelated version I've created I'm going to have this one inside a Gameboy screen for my business cards.

Thursday, 17 April 2014

Architectural Renderings: Two Point Perspective

The first task was to sketch a building from life and present it using techniques commonly seen in stylised concept sketches, for use in product, building, and vehicle design. This would give me the opportunity to practice not only my sketching but also my presentation skills. I've always been keen on perspective drawing I feel it fits with my perfectionist inclinations and apart from the few sketches I did for my vehicle design module I haven't had another opportunity to practice.
Looking at the way other designers present their work (such as Syd Mead) we can see in the images below that these early concept sketches concentrate on line work with some simple shading to signify light direction and enhance the 3D appearance in conjunction with the use of perspective.

Two point perspective

This technique is great for producing stylised, 2D illustrations of objects to make them appear more three-dimensional. When viewing an object in real life it seems to shrink as it gets further away, parallel lines will eventually converge into a single point on the horizon/eye level known as the vanishing point
I took plenty of reference photography of buildings around Newcastle that I thought would make for interesting sketches. Modern buildings with strong lines/form would translate well to a perspective sketch.

I've always loved the Baltic and thought it was a rather appropriate subject matter for the work I wanted to produce, not only in form but also the use of the building. I sketched the image by eye using the photo only as reference but as you can see in the image below I've shown where the horizon line and the vanishing points would be on the building from the viewers point of view.

I'll generally start a perspective drawing by establishing the horizon line, then the longest, vertical line that is closer to the viewer, diagonal lines drawn from the top and bottom of this line to the vanishing points act as constraints that all other lines will fall between. I'm really pleased with the results, I took great care to capture the lines of every detail on the building which greatly enhance the final sketch. It's probably a little stretched at the top but I feel this adds to the imposing nature of the design.

To ink up my sketch digitally I decided to use Sketchbook designer, a vector based version of their sketchbook pro software. I setup the perspective grid in the program and redid some of the line work from my initial sketch as some of the vertical lines weren't straight, this somewhat fixed the stretching.

I continued to refine the line-work and shading, adding all the window details with increasingly thinner lines. In my reference image a window was open so I added that to my design, it's this little detail that grounds the building in reality. Once I was happy I took the image into Photoshop to add the text to the building by skewing and distorting it until it followed the perspective I'd established. I added shading to each window to emphasise the light direction of the image whilst experimenting with a variety of brushes to create the light streaks I wanted.
As a finishing touch and to make the building 'pop' I created some background noise to frame the building. It really helps the building stand out but I also tried to suggest shadows, and even clouds in the noise that I created.

Tuesday, 15 April 2014

Weekly Targets

  • Collect reference imagery/photographs of buildings in town to sketch
  • Create my personal profile for the college 'Look-book'
  • Collate and update 5 images that best represent my work
  • Start working on a logo/profile image

Monday, 14 April 2014

SWOT Analysis

We're coming towards the end of the year so it's time to reflect on my skills, what I've learnt, and where I can improve to make me stand out from the crowd. I already have a solid foundation in 3D modelling and this year I've filled in some of the major gaps I had in translating those skills to game engines, such as UV mapping, texturing, and normal mapping. I've already signed up for my third year, to top up my foundation degree to a bachelors, as I believe it's needed to refine what I've already learnt. This should also allow me to get a head start experimenting with the return of virtual reality and the new design/market opportunities that I believe it will bring.

Using the site I can compare what my strengths are in relation to a specific vocation and what I need to work on to be a more suitable candidate for that role. As much as I'd love to break into the game industry I certainly don't want to limit my options, and I believe a lot of my skills are transferable across many creative industries.

My current skillset:

3D modelling: This is probably my strongest skill, I can take almost any design and translate it to 3D. I've focused on topology and optimisation to make my assets 'game ready', for animation and game engine implementation. I'm proficient in 3DS Max and UDK, but I need to work on organic sculpting software such as ZBrush or Mudbox to complement my toolkit.

Texturing: I've had a lot of experience this year with unwrapping UV's and texturing my models and I feel I'm progressing well, although still need more practice. Baking high resolution details to diffuse and normal maps was one of the area I felt needed covering after my first year, and this year I've had a chance to experiment. As with my modelling, this area would benefit from continuing with my Zbrush/Mudbox self directed study.

Game Design: One of the greatest skills I'm taking away from college is the process of design from conception to realisation. Given a theme or a brief I can create character, environment, or new gameplay ideas through the use of thumbnailing and silhouetting techniques, and carry that through to fully realised 3D assets. Being passionate about games for the last 30 odd years means I have a vast knowledge of gameplay, mechanics, current trends, and the industry as a whole, but I'd like to focus on areas such as ludological studies/game theory in my third year with the potential to build my own complete games from scratch.

Storyboarding: I've really enjoyed projects where I've had a chance to create my own storylines for my game ideas. I find it easy to visualise my characters and environments and the actions they may take when placed in specific situations, but also how to setup 'cinematic style' camera angles and movements to capture the action. As well as writing screenplay like break downs of my stories that anyone could follow, which break down scenes into manageable chunks of animation, I've loved creating my own animatics based on my concepts. During the summer I need to finish the opening cinematic of my 'Illuminated Knight' game that should compliment my overall design, but has been put on hold due to lengthy render times.

Character Design: I've excelled at creating fully realised characters with each opportunity I've had. I enjoy the conceptual stage, researching, creating mood boards, and silhouetting, to give me a strong foundation in realising my characters with biographies, traits, look and style. I'm not going to kid myself, there are much better character designers out there, but I certainly believe I can play a role in conception, translating ideas to 3D models, or communicating my ideas to other members of a team.

Vehicle Design:  Another strength of mine especially with my hard surface modelling skills. I'm a bit of a perfectionist with a keen eye for detail and have always loved creating highly realistic mechanical designs and renders. I've only had a couple of projects based around vehicle designs but it's something of a hobby of mine, so will continue to create my own portfolio work in this area.

Environment/level design: I've had a lot of opportunities this year to work within game engines creating environments and levels. As well as conforming to the technical constraints, it's pushed me artistically.

Concept art:  I've argued that the course should be renamed  'concept art for video games' rather than Game Design as this is what we've covered most over my time. It's certainly helped with my creativity, imagination, artistic ability, and conception, but I still feel I'm a long way from doing it as a career. It's a really competitive field with very few secure positions within the industry, with most concept artists being freelance or hired for the initial stages of a project.

Video editing: I've loved the time I've had learning After Effects and the animations and videos I've produced using it. Timing, pace, and movement have all been areas that I've been strong in

Extensive hardware knowledge: This is a no-brainer, I've had a long history with computer hardware, being obsessed with computers from an early age, and have built my own PC's for over two decades. The hardware has also fuelled my passion and ability to work in 3D but also to experience all the latest games. I'm always looking for the next big revolution with each computer or console iteration and is why I'm so excited about the return of VR. I've also worked as a computer technician building high-end gaming systems to order and know what each level of hardware can do in terms of performance in relation to games. I could easily fit into any studio as a computer tech, troubleshooting not only hardware but software related issues.

Team management: In almost any job I've had I've started at the bottom, worked hard and then found myself in positions of responsibility before I knew it. That has come from being a strong team player, and whilst I'm happy to work alone, I like thinking outside the box to implement ways in which we can all work more efficiently as a team.

Lots of patience: While it's not a conventional skill it's something people have commented on in both my personal and professional life. It's a trait that helps keep me calm and focused under pressure but also for when I'm stuck on a problem and persevere until I find the solution.

Potential Job Sectors and the relationship with my skillset:

Game Designer: Game designers create the initial concept and vision of a game. They create the world, setting, story, characters, and all other game elements and design details, and then communicate this to the rest of the development team who create the art assets and computer code which allow the game to be played. I really believe this is the area I would excel, hence me desire to focus on the subject at college. Having managed multiple teams, a keen eye for detail, the ability to bring the best out of people, a good communicator, and extensive gaming knowledge, I believe I'd be perfect for the role....eventually. The problem being it's not an entry level position, you'd have to have a proven track record or worked through the ranks of a studio, because the reputation and financial future of that company rely on your decisions. It's something I'd like to work my way towards, but another route, with a great game idea under my belt, could be to crowdsource funding and set up my own studio.

Creative Director: Very similar to the game designers position, although probably in a studio or on a game with a number of designers working on a large project, the creative director will liaise with all the other designers and split the project between them. With my team management and communication skills, I think it's a job I could be good at, however similar to the game designer position it's not an entry level position.

Concept Artist: The concept artist will take direction from the game designer to produce the initial artwork for a game based on their ideas. This can be broken down into further areas of expertise such as character, environment, and vehicle design. The look,style, and mood of game assets are then based on those initial concepts. These pieces can help 'sell' a game idea, and as well as the feedback from the game designer, keep the team on track throughout the project. The concept artist skillset will cover silhouetting, thumbnails, and digital/matte painting. Often concept artists are only utilised for the early stages of a games development so it's not a secure/stable career, with work often being freelance or outsourced to studios that specialise in the production of concept art. There are so many talented concept artists in the industry and I don't believe my skills are strong enough to enter that field at the moment. My 3D skills would be advantageous in this field for blocking out forms to be overpainted however taking 3D models further requires a lot of behind the scenes work (optimising and unwrapping) which doesn't fit with the rapid development that's the nature of this field.

Level Editor: In the early stages of development the level editor will block out the major forms, and pathways of the game environment, setup objectives, and plan the layout of items and NPC's, based on the initial concept work and feedback from the lead designer. I've had a bit of experience with UDK now, planning and layout is intuitive and you get immediate feedback with what you've just built by jumping into the game in a preview mode. With more practice and completing the levels I've started to work on I believe I'd have the skills to enter this position at a junior level.

3D Modeller: With my current skillset I feel this is my best route into the industry. The Modeller will create all the 3D assets for a game based on the initial concept art. I'm really strong with the technical side of modelling, having a good understanding of topology, optimisation, and UVW mapping, so would relish the chance to work on 3D assets for any project. However I've let my organic modelling/sculpting slip so I doubt I could be the main character modeller without more practice, especially as the software itself (Zbrush and Mudbox) has evolved so much since I last used it.

Technical Artist: The technical artist bridges the gap between the artists and the programmers. They understand the technical limitations of what can be achieved within a game engine and how to implement the vision of the designer and artists within those constraints. They have a good understanding of all the software involved in the game development process but also research new types of tools/software that may enhance workflow and then communicate that to the team.
Being more technically minded I feel this is another viable point of entry into the industry for me. I have extensive knowledge of the software involved in game development but also the hardware on which games run. I'm constantly experimenting with new types of tools to improve my own workflow, and have found the more I experiment the more transferable skills I have with each new piece of software.

Quality Assurance Technician/Tester: This is probably the best route for entry into the games industry, for some one in my position with no previous experience. It's a junior position but it gets your foot in the door and then there's room for progression, but also the industry experience I need so much.

Areas that need work:

  • Work on my 2D sketching/illustration concentrating on anatomy
  • Need more practice with organic 3D sculpting software such as Zbrush/Mudbox
  • Ludological studies, study game theory and the mechanics of gameplay
  • Motion graphics, need more practice working with 3D in combination with Video editing
  • UI design, need to work on UI elements, layout, and player feedback
  • Create more concept pieces using 3D but then overpainting
  • Speeding up my workflow
  • Refine my Portfolio

Plan of Action:

  • Finish/refine the artwork I've produced over the last couple of years
  • Create a new game idea using character silhouettes and environment thumbnails. This will help me practice my 2D design skills, implement 3D work into 2D illustrations, and bulk out my portfolio
  • Using my new designs, practice speed modelling in Zbrush/Mudbox
  • Purchase and play different board games to study mechanics/game theory

Friday, 11 April 2014

Practicing my ZBrush Skills

To continue my Zbrush practice I found a more detailed tutorial that covered the creation of a forest elemental. Using Zspheres I roughed out the initial form of the body.

Once I was happy with the body I moved the Zspheres to create a dramatic pose for the creature.

I converted the Zsphere pose to an adaptive mesh and sculpted in some forms using the clay buildup and move tool, it's much easier to make big changes at this stage when the density of the mesh is low. This mesh acted as an armature on which I used Zsketch drawing random tubes from the chest and over the shoulder to build up some mass that would be used as a cluster of stones and branches. Converting this zsketch to a mesh retained the individual polygroups which I then inflated from the connecting 'mortar'.

With the mortar masked off I started using the Hpolish and clay buildup tool to shape rocks and branches in the polygroups that suited the form. I tried to make it look as if the branches were wrapped around the stones binding them into the mass.

Using zsketch I built up some mass on the other side of the armature. I used the mesh insert tool to insert a tube that protrudes from the shoulder and shaped it into the stump of a trunk, enhancing the feeling that the creature is cobbled together from random parts of the forest.

For the head of the creature I duplicated the armature and deleted all the unnecessary zspheres leaving just the the head section. After converting this to an adaptive mesh I blocked in the rough forms and positions of the mouth and eyes at a low resolution, helping to establish the initial layout of the face. I wanted the creature to look angry but also in pain, annoyed and puzzled at the misfortune of his own existence. I used the claytubes tool to insert branches that look like saliva as they spills out of his mouth and cling to the side as he roars as if the side of his mouth is tearing open.
I spent a good few days going through this project and he's coming together nicely, however I felt I'd learnt what I needed to sculpt my game character and I had to move on. It would probably take another week or so to finish this guy off so it's something I need to come back to when I've finished my assignments.

Wednesday, 2 April 2014

Real Time Path Tracing

Game engines are continually evolving, and with each generation they try to improve rendering techniques, the ultimate goal being  graphics that are indistinguishable from real life. Ray tracing simulates light in a realistic way by tracing each photon and how it interacts with a surface, but has only been available in 3D modelling programs with lengthy render times as they calculate each bounce. However with the brute power of today's graphics cards real time raytracing may be just around the corner.
In a similar way to how my favourite rendering program Keyshot works (hence the noise as the camera moves) this new engine uses 'Path-Tracing' to bring realistic lighting in games that step closer, it's pretty impressive.

Wednesday, 26 March 2014

Facebook acquire Oculus Rift

I've been really excited about the return of VR since the hugely successful Oculus Rift Kickstarter campaign a couple of years ago. With each titbit of news my excitement has only grown, and I plan on getting one as soon as it's released. It's something I want to specialise in developing for in my third year at college as I really believe it's the future of gaming, and should open new design and marketplace opportunities. I guess I'm not the only one with today's announcement that Facebook have purchased Oculus for $2 billion. As much as I'm not a fan of Facebook I'm happy that Oculus now has a huge company behind it, the future of VR is pretty much assured now. However I do feel sorry for the initial investors, they invested in something that they felt part of, something that would grow organically and be championed by enthusiasts and then maybe becoming mainstream. I worry that if not handled properly the return of VR may be a flash in the pan like 3D, and if these big companies don't see an immediate return on investment VR will be considered a failure to be shelved again for the next few decades.

Friday, 21 March 2014

Unreal Engine 4 - Features Trailer

After getting to grips with the fundemantals of UDK3 in my last few modules, I'm excited to see the improvements and upcoming features that this trailer for Unreal Engine 4 showcases. Hopefully I'll be able to enchance my workflow, add an extra level of realism and detail to my designs, and keep ahead of the game with this new toolkit.

Monday, 17 March 2014

Weekly Targets

  • Research for job opportunities in relation to my current skillset and where I need to improve.
  • Find a tutorial to help me practice my Zbrush tuition.

Friday, 7 March 2014

Creating Game Quality Vegetation Assets

One of the things I was keen to add to my level was vegetation, I thought adding these elements would really bring it to life, fill in bare areas of the terrain, and add interest to the landscape. The problem with creating these elements was in keeping the polygon count low while making them look as realistic as possible. Trees are also some of the most complex natural shapes and whilst there's plenty of ways to create trees, including presets in Max and programs like SpeedTree in UDK, It's a technical process that I wanted to learn, and update my skillset.

With this being a very technical discipline, to create the vegetation I had to follow a tutorial that covered this process. I found a great tutorial by 3D Motive called 'Creating Foliage for UDK' that I used to help teach myself how to bring these elements into UDK but also animate them though the use of materials in the engine.
The first element I learnt to create was grass, again what should be something simple, if we created individual blades of grass as models and then import them into UDK, we would need thousands of instances to get any decent coverage, and the polygon count would be astronomical.

To keep the poly count low but still create realistic looking grass, we can still model the grass in a 3D program such as Max and then 'bake' the models to a texture plane. In this way we can capture all of the lighting, specular, and diffuse information of a 3D asset but then it will still read in a similar way on an image plane when in a game engine.

Once the texture was complete I then took it back into Max and applied it to the plane that was used as the render to texture object. I then split the plane into the two smaller polygons at the top and the larger grass asset at the bottom. With preserve UV's enabled I then moved the vertices of each asset to tighten up the geometry around the grass texture, it also means less geometry is rendered in UDK (every little helps).


Both the grass and trees were going to be animated in UDK through the use of materials. To do this I had to paint the vertices using 'vertex colour', a red gradient lighter at the top meant the grass should bend more at the top whilst staying fixed at the base.

I went through each asset in a similar fashion and then created random clumps that I can scatter about my terrain.

 The trees were created using the same techniques as the grass, modelling the branches, baking textures and normals, painting the vertices, and then grouping the planes.

  I used splines to create the trunk and branches, and then attached the planes to them trying to create a naturalistic looking tree.

Tuesday, 4 March 2014

Weekly Targets

  • Continue working on assets for my Knight level
  • Choose some of my best models to retopologise for my art gallery
  • Update my online portfolio

Tuesday, 25 February 2014

Grounded: The Making of The Last of Us

Sony have just released a full length, behind the scenes documentary about the creation of Naughty Dogs' 2013 masterpiece, The Last of Us. It's really interesting to see the development process, and studio workflow behind such a big title. I've now heard, quite a few times, how big projects like this tend to come together right at the last second, John Lasseter said that all the Pixar movies were on the constant brink of failure, and last year at Gamehorizon David Polfeldt (Massive Entertainment) said that Far Cry 3 wasn't great until the last stages of development. So it's refreshing to hear Naughty Dog say the same about The Last of Us, it's similar to the way I feel on each module!

Monday, 17 February 2014

Creative CV: Art Gallery Portfolio

I've had to put my 'Illuminated Knight' project on hold, I've been asked to update my portfolio as people at college are interested in seeing what I can do. Morrisons have contacted 'the powers that be' at college enquiring about the possibility of making a 3D walkthrough of one of their stores, with the work I've done so far in UDK my tutor has put my name forward. I wasn't happy with showing them a walkthrough of my level in its current state, so I thought I could kill two birds with one stone by creating a 3D art galley in UDK that could act as an interactive portfolio which I could send to prospective clients. I feel this would be a great way to showcase models, videos, and images that I've created rather than just a list of images on my blog.

After sketching out a quick design I began blocking out the gallery in UDK using simple BSP brushes. The viewer will start in a large atrium/gift shop that will have details about me, my interests, and skills that will serve as a more traditional CV. From the atrium they'll be free to explore the gallery as they see fit with rooms devoted to each project or a specific area of design. This will be an ongoing project as my portfolio develops, but I've left plenty of room for expansion, so I can add new rooms for each project I work on. Once I've created a few of my own game levels I want the viewer to be able to click on images of said levels and be transported to them (ala Mario 64). Eventually I want to add NPC that walk around and comment on the artwork within the gallery. Maybe I can voice a grumpy old man pointing out the flaws with my own designs?

I started to apply some simple textures, trying to create that modern gallery feel in a refurbished old building, as with real galleries the building should not over-shadow the artwork but enhance it.

This project was another learning curve, I was having issues with shading and found BSP brushes behave differently to static meshes in UDK. To get better shading on a static mesh you have to increase the resolution of the shadow map, but it's the opposite with BSP brushes...who knew? It was also an opportunity to experiment with materials and shaders, I'm quite pleased with the chrome material I've made for the metal posts.

Doing this gallery has taught me a lot and will not only inform projects in the future but should feedback into my 'Knight' level. One of the great things about college is the ability to make mistakes, and learn from them. Making mistakes at work is a huge no-no but at college I've had the opportunity to experiment. Unfortunately I didn't realise that when you preview your level in UDK your character is 120 units high, however in game it's only 96 units! This means that the scales in both my levels are wrong and would need some major reworking! It's a mistake I'll never make again.

I've still got a lot of work to do on this gallery, mainly descriptions and plaques that explain each project, but also sound effects, menus, overlays, and some interactivity for the viewer. As I said it's an ongoing project so I'll be returning to it over the next few weeks.

Monday, 10 February 2014

Illuminated Knight Level: Work in Progress

Having made the building models and terrain, unwrapped them, and begun texturing, my level is starting to come together. There's still a huge amount of work to do and at the moment the stained glass effect isn't obvious, that should come when I enable the edge checks renderer, and add my own custom textures to the sky and ocean. However the tight Medieval feel of the streets and the large cathedral structure are beginning to look like my initial designs.

It's been a huge learning curve especially as I've mainly been left to my own devices with no tuition on the software or even basic level design. However with my own artistic ability, perseverance, research, and hard work I'm starting to piece together the work flow involved in making a game level. I realise working in a studio I'd specialise in a specific area but this project has given me a great overview of the process, and really makes me appreciate the Herculean effort that's involved in creating games today.

Saturday, 1 February 2014

Illuminated Knight Opening Cinematic

Once I'd rigged my character and created the cathedral interior it was time to bring it all together. Based on my storyboard I broke the scene down step by step to help make the animation more manageable.
The first time we see the knight is when he opens the cathedral door and a battle rages behind him. I'd already created vectors of knights fighting to animate in the background and wanted to blend them into a fiery background to emphasise the hellish world the knight was trying to escape.

Light is very important to me in this scene as it plays a big part in the overall theme of the game concept. As the knight walks down the aisle I wanted the light from the stained glass windows to ripple over him, almost as if it's defining his form. To achieve this I used volumetric lights with the stained glass textures as projection maps. It's the first time I've used volumetric lighting and I had to experiment with different settings to achieve something I was happy with. I animated the phase value in the volumetric fog settings to simulate small dust particles that move about within the light shafts, to suggest the building has been neglected for quite some time. However the more I added to the scene the more difficult it was to get feedback on how it was looking as preview renders were taking forever!

It's way too bright at the moment but there's still a lot of experimentation to do before it's finished, but as proof of concept it's starting to come together. I've set up the various camera positions, but still need to work on the animations of the knight before I can move forward.
 If I'm honest it's been a bit of a struggle, I came to college with some knowledge of 3D modelling and hoped my tutors would be there to help when I needed answers to specific problems with rendering, animation, and lighting. As self motivated as I am and the countless hours I spend searching the net for answers when I get stuck, the college has not been able to help with any of my issues. At the moment I'm struggling with render times, it's going to take days to render this animation as is and I just don't have the time to take my computer out of action for that long. If college set up Backburner on the networked PC's they have 3DS Max on (something that would take minutes) I'd be able to use a render farm and get it done in a night, but throughout my time I've encountered heavy resistance to any kind of change. For now I'm going to have to leave this animation and come back to it during the summer when I can render it in my own time.