Major Project Evaluation

My major project was again entirely collaborative this year. This was a double edged sword in some ways, as it meant I had to manage time to fit in with my colleagues schedules. The upside was that it was really fun, and it allowed me to contribute more to making the project a success. The four projects I chose fit well with where I want situate myself in terms of skills. The rigging project, while only as quarter of my total workload, was very intensive. Having to produce two rigs of sufficient quality in a restrained time really helped my workflow, and then being able to animate with not only my own rigs, but my colleagues gave me insight to the strengths and weaknesses of both. The Unity block sat well with my rigging skills, as I was primarily concerned with animation and character setup and AI. Overseeing the process from rigging to implementation gave me an insight into how to plan and execute game projects.

 

Rigging
Overall, I felt the rigging module went well. I managed to create two rigs in four weeks that the animator's seemed to respond well to. There were features I would have liked to add, but did not have time, but I tried to prioritise feature additions by how much benefit they would give the animator. The main problem with the rigging module for me was a time imbalance. The dragon rig took longer than the Fae rig, which I had anticipated; the dragon was a more unique character with a physiology I had not had experience rigging before, whereas the Fae was a biped so I was more within my comfort zone. This made the Fae a little bit rushed, and I would have preferred to have had a little more time on the rig.

The main areas I would have liked to have been better were to do with the staff system. I knew this would be a challenge, as it required the IK arms to be very robust, so I made sure there would be stretching, and the ability to lock the hands to the staff. The staff can also lock to the characters back. These functions were fit for purpose on Echoes, but because they were created via parenting rather than using script expressions, there were areas where the system was weak. The hands had to be at zero position/rotation to stick to the staff, which meant interpolation could get messy if not controlled  Ideally, I would have better parenting and keyframe controls held under one simple UI.

Usability is another area I would like to be better. A number of times I encountered animators not using features because they simply did not know they were there. I provided read-me files with explanations, but a text document is hard to remember when getting on with animation, so I want to look into having switches like IK/FK, global/local, and selection sets available from a custom UI.

Unity
The Unity project went well, and we achieved a great deal on the project as a team. We pushed the project further than we thought we could, and though the final game is a little rough-edged, it is in working order ad the elements are all functional. We knew from the beginning that as none of us are programmers it would be difficult to build a working game, but I think we overcame most of the hurdles we encountered and found solutions to most of the problems.

Playmaker was a huge advantage to us, as it simplified the scripting process, particularly when dealing with animations. We still found ourselves using a lot of javascript, but this would have bn much more complex without combing it with playmaker's state machines.

Being in charge of the AI, I had to come up with very simple solutions to very complicated problems, and have it look fairly convincing. AI is notoriously difficult to program, and ours are incredibly simplistic. However, I think that they work reasonably well, and though gameplay feels very much like a game, with little sense that the characters can think, the combination of the AI we have and the art and animation really helps the characters to feel alive.

The player controls were something we wanted to try and get very solid, and I think overall they work well. Occasionally there are areas that could require more polish; when running down a steep hill the character will continually fall small amounts at a time, which is jarring for the player. Overall though, I think the game feels responsive and functional, so I would call the unity project a success, and a massive build on the work I produced for mongols vs ogres last year.


Animation
I love animation, and it is often frustrating that with a more generalist skill base (unity and rigging taking up 50% of my time) I have less time to work on my animation skills. For Echoes, I think there are mixed levels of quality to the animation, depending on the character and the animation and the time I had to put them together.

The bark basher was challenging, but very enjoyable. The character had a distinct personality, and being a huge boss character, it was great to see his animations in game. The heavy shields provided two interesting problems. First was animating them convincingly whilst still giving the Bark Basher enough mobility to effectively fight the player. Second was making sure that we animated his shields at a consistent weight throughout all the animators. Whilst it didn't matter for showreels so much, in game there would be no distinction between animators so it would be jarring for the shields to switch weights all the time. The weekly meetings really helped us to nail this down and kept everyone working to a similar style.

The Fae was reasonably straightforward for the most part, being a standard biped. The staff I felt could have benefited from better controls, as I struggled when trying to pass it between hands or spin it a great deal. The  main staff control had a tendency to gimbal lock if rotated too much around the x-axis, so the axis could occasionally get skewed. Many animations suffered from jarring endings due to the need to get back to the idle pose, but this is less noticeable in game so it was not a major issue.

The Dragon was great fun to animate, and the only real issue I had had was with the FK tail, it was difficult to get it smooth, though turning on ghosting helped. After I added the dynamic tail it allowed the final layer of polish to sell the animation.

I only had one animation for the centaur, which is a shame because I found him a lot of fun to animate. Te dynamic between his horse lower half and human upper half made him feel quite primal, and it was a fun challenge to try and get that across in my attack animation. The animation was a little locked down to one axis and I'd have preferred to get the centaur moving around laterally more, because in the final game they can appear a little direct, which doesn't leave much room for showing off the artwork.

The animations for Genevieve had a good base, I thought, but needed refinement  Here animation for playing on the ruins needed more cleanup and smoother motion, and more exaggeration when she almost falls. The animation is tiny in the game though, as she is so far away, so I didn't clean it up as much as I should have. The wave animation is very rough, and didn't make it into the game, but I think it had the foundations of a decent animation. The animation for the outro cut-scene went well I thought, though I didn't have time to render it off at full quality before the hand-in. I got the particles and smoke effect set up, which was a lot of fun though, and the final version will get done at some point soon.


The project went well overall I thought, and though there were plenty of hurdles and problems, I felt that they provide me with clear direction as to where to invest my energy into the future as I improve.

Website is live!

I got round to making a website at last, and its now live at joehornsby.co.uk. Its still lacking a proper showreel, as I'm finishing major project next week, so I'll be able to put together a better reel then.

Also recently I tweaked the dragon rig to have a dynamic tail. Here's a peek at a test I ran with it:

Fun stuff, there should be tons of new stuff in the coming weeks as we come to the end of major project, so I'll post up as much as I can.

Dragon Update

Not long left now, and I'm just posting up a quick update on the dragon animations. I've restructured the claw attack a lot, and I think it works a lot better like this. The attack is anticipated better, and it has greater energy now. The end needs a bit of tweaking, as the first flap feels very unnatural. This is close to done though, a few tweaks and another pass at the tail and I'll call it finished.

At the suggestions of our tutors, we have added a big dragon to the end of the summer level, to act as a level boss. This means we need a few new animations, and fortunately a few people have agreed to take on a couple more. I've taken on the death animation for this big dragon, and I've been playing around with it. Here's what I have so far:

I like the feel of it, though I need to play with the timing and the weight of it. There's some jerky movement in the wings as well at the start that needs to be ironed out. Overall, its a little melodramatic and Shakespearean, but I think from the low front angle it should put across the size to the player.

Inspiration: Valve

Valve are one of the biggest game studios in the world. Often described as the 'Pixar of games', they have a long history of producing critically acclaimed games, and are beloved by the game community. Valve are also  famous for their work environment and ethic, with all their employees working willingly on projects of their choosing. 

They built their name on the Half-life series, on which their pioneering work led the way in terms of narrative and storytelling in games. They were amongst the first studios to realise that game characters could give a performance that elevated the story and added great depth to the gameplay.

They went on to create much loved recent successes with the likes of the Portal and Left 4 Dead games, not to mention the multi-player shooter Team Fortress 2. All of their games have a versy strong art style and solid design, with a focus on the player as the most important aspect of their game.

Valve is a giant in gaming culture, and along with the massive commercial success of their Steam PC gaming platform, they have kept to their roots, favouring effective game design and narrative over rushed release schedules. The company is renowned as a fantastic work environment, and their ethics transfer well to any team effort.

Bark Basher Damage

I've worked on the Bark Basher heavy damage animation a fair amount since the last time I posted. There's now a section at the end in which the character spins around to face the player. This was added in as the second part of the animation will be triggered when the player is behind the Bark Basher, attacking from the rear. The animation still needs some clearing up, I'm not particularly happy with the initial reel back, it feels a bit jerky, but overall I think this looks a lot better than it did.

First Submittal

Time flies, and time you need to work on something even more so. First submittal has snuck up on us, and we're all trying to compile our work. This is our last chance to get the tutors to have a really good look at our work, and our last chance to get some in depth feedback.

Most of my focus was on the rigging I've done, as that will go (mostly) unchanged between now and the hand-in.

The reel is missing a few features, ones that I forgot to mention if I'm honest. I haven't made mention of the variable parenting for example. Its long and ponderous, but I developed it mainly for the marking reel for uni, rather than my own showreel, though I will use elements from it in my final showreel.

Inspiration: Rockstar North

Rockstar north is one of the oldest and most prestigious games development companies in the world, as well as being the largest in the UK. Their Grand Theft Auto series is amongst the most popular in the medium, and they meet with critical acclaim upon every release. The studio was founded in 1988 as DMA design. Eventually the studio grew into Rockstar Games, and as they expanded and opened up other studios, the Edinborough studio became known as Rockstar North.

The Grand Theft Auto series is infamous for its violence and controversy, but also famous for its technological advances. As a frontrunner in open-world gameplay, Rockstar is often at the centre of gaming news with advances in engine technology which allow for a greater level of immersion.

Relatively recently they included a real-time character physics engine called Euphoria in Grand Theft Auto IV. This piece of technology seamlessly blended the animations of the team with physics driven motion, to create stumbles, falls, and allow previously complex motions like climbing steps to be more easily achieved.

Above all this, Rockstar values storytelling as their most important asset. This is key to their success in recent years. Their writing and animation is cinematic and pulls few punches on subject matter, painting a stark picture of the lives of its protagonists. Their games blend gameplay and story into a smooth whole, and I would love to be able to work on similar projects myself.