Week 30

Extended Reflection

Currently I am studying Game Art Design at DMU, this is in the hope that when I graduate I will not only be a better artist, but be employable for these skills.

My main aspiration is to become an Art Director or Lead Character Artist. However I have noticed some gaps in my arsenal, namely software like Maya or Z Brush. These are both software used regularly in industry that I have a very simplistic understanding of and I shall need to hone my skills on such to become more employable.

For instance, Blizzard are a renowned company that require both Maya and Z Brush as Proficiencies for a lead character artist position. Working on a stylized but semi-realistic title like Overwatch is a very real aspiration for me. Looking at my current standard of work, the hand painted texture look has been successful in my projects, whereas the more traditional photo realistic or grunge textures have been less effective.

My current standard of texturing has shown great improvement from where I was at the start of the year and my iteration is more focused, quicker and shows greater understanding of strong silhouettes pertaining to which of them I should advance. Because I have been able to get through this stage quicker, I’ve been able to paint over my work and study the compositions I’m making in engine closer; allowing a higher level of polish on my work.

I now know pushing my visual skills in Photoshop and my understanding of the fundamentals will ensure I am T shaped artist that fits movements in current industry. However there are visual skills I will need that must be learn off my own back to help me, painting over work to show my meaning and refined concepts are both things I need to work on off my own back in the coming year.

Concentrating on the skills I will need in the coming year rather than just the ones I have developed and refined while on course, I must become more confident and outsource my feedback on sites such as Art Station and Polycount. Without a larger, harsher feedback pool I will not progress at the rate necessary next year. This is because my finial year requires me to strive into industry slandered and achieve work that is worth putting into portfolio.

There comes a time when I have to accept that my head is full of tangents and loops, it just wants to tell you so badly about all the things that it kind of mushes them in an incoherent jumble of consciousness. This year I have learned how to decipher this illegible mess my brain wants to give you and place it into a somewhat coherent blog. Although in honestly the writing has gone well in comparison to last year, the confidence to post said writing has not gone well. Unfortunately I keep leaving myself chunks of blog to edit. While my university helper tries to help me decode my nonsense and is starting to feel real improvement when it comes to how coherent my work is; I personally do not have the same sense of achievement.

Looking back there wasn’t really a lack of time this term but last term I really struggled to find time and it is these blog posts that have suffered. If I was to redo the blog I think I would post smaller more summative posts and make them only around 500 words maximum each. This would mean I would be less nervous that my rambling where being read and my blog count would triple.

Although the blogs haven’t gone as well as hoped one major thing that has happened is I feel able to write by myself these days, something most people find easy but up until around two months ago was both scary and seemed implausible due to my lack of ability in writing and proof reading over my work. Have really helped me find my own way of reading my writing over and having more confidence. I’ve been trying to get small feedback from peers and this has helped me see the small improvements in my writing that are happening. 
Moving on to what has supported almost every project this year, my 3D has definitely come on leaps and bounds. Having spent some time over summer looking at characters, due to my interest in them, I was able to model things more coherently. I would say my current standard of modelling has now developed past coherence and is pushing into being proficient. Admittedly the things I have been most successful with have either been props or characters, which leads me to reflect on what needs to improve next.

I think one of my jobs over summer will be to start looking back in to engine but this time with focus. Because engine offers so much I think I should really hone my research and learning to aspects that will aid the parts of industry I want to go into. As such: particle effect, engine textures, importing characters, creating simple NPC animation loops and changing out the dummy for my own meshes are all skills in must invest time into learning.

One type of task that can be difficult to reflect upon in hindsight is group work, I can honestly say that my group work has been a very mixed bag. On one hand it went well for some of my first projects and the groups I worked with for the most part allowed me to reign in ideas and act as an art director of sorts. Most notably the Alice project where many assets are put past me before they enter engine; supporting this a lot of the consepts are painted over by me and the theme, atmosphere an overall look of the level where pushed by myself. I find this really exciting seeing it come to fruition and I’m very proud of the work that we’ve produced thus far. Here’s to hoping it is finished and works as well as well envision it working when it’s finally all in place.

 Even though my group work hasn’t always been fabulous I feel that having one good one bad and one downright ugly project is the best way for me to learn how to learn the skills needed for art direction. I need to know what not to do and I need to learn those lessons fast. Where better than, at university where I have a nice safety net for when it all goes wrong.

Looking at all the parts of my year and how I have advanced I can easily see where it is I need to advance further and where I need to balance my skills. My first plan is to use the summer to gain a better understanding of engine, character design and modelling and to really push my visual skills so they make me stand out as a strong candidate for future careers. Here I’m thinking of my hand painted textures, my Imagineering skills and my understanding of fundamentals. People need to respect you for you to lead as I wish to and there is a very simple way of getting that respect among artists, it’s quite simply to be really good at what you’re doing. In the third year I will create characters when possible and really push to direct projects where possible, focusing on painting over work and giving good quick feedback to people I work with. I will also focus on learning software I will need for the next stage, like z brush, 3D coat and Maya. This will not only aid my work and make it look better but it will make sure I am proficient in software used in the character roles I am looking at applying for after the course. After all if university is my time to do nothing but learn then I shall endeavour to learn more than just what is set out for me to learn.

Week 9

Concepting and Modeling in sync

Final idea tonal 

final idea tonal

I wanted to use the fact that I had already got two strong ideas completed early on to my advantage, For this project i would model both characters as my stretch goal.

Ideas for patterns

Ideas for patterns

Colour concepts

Colour concepts

It wasn't long before i had the first attempts underway but i quickly found that the shape of a bell, or in this case poncho is not an easy shape to pull out of a cube. This led me to try many other shapes to find the easiest way of forming the ponchos for my two characters. In the end it turned out I will forever be the box modeler I started off as. In order to have nice quaded and looping topology, starting with the good old trusty box, is the only way I could make them.

Once the great poncho shape debate of 2014 was over it was time to move on with my life... I mean models and tweak the original designs to fit a little better in 3D.

This week has now ended in unwrap city, but texturing them is next weeks ordeal, good luck next weeks me...

Week 8

Character Project

I start this project on a high. My seminar on the dichotomous duo was really interesting; the part that was most informative was when we wrote an alphabet of words on a board as a contest. Then we had to find a dichotomy with one of these words and make a Venn diagram.

I chose sour and sweet as I had this image of an old divorced crazy lady and a little cute girl on her paper round, having a little bond over cats... Because cats are great. However because I felt that this could go into a colour palette and theme too similar to my sentry gun I changed this to leather and lace.

My investigation into a theme has led me down a tribal route. Looking into Peruvian and Mexican culture, high fashion and illustrative styles I could base my characters off. I found someone called Yao Yao and it's great, i'm in love with the skinny leg bulbous body look. I'm genuinely excited to use their work in a style guide and see how it converts into 3D.

It's funny how solid ideas repeat, even if it wasn't intended. I looked at journey because the more I looked under my theme and style the more it popped up. Low and behold the strongest of my silhouettes where visually similar to the shapes in journey. It may have been unintended but it does show how there is a defined difference between ripping off something and riffing off something. By using similar source imaginary I came to the same conclusion, and rather then panic and wonder what to do, I decided that this was great. I had found a style, a theme, made strong choices and produced strong designs. The fact that they are akin to something often seen as very visually strong can only be positive. This was the birth of Journey 2: Electric boogaloo, with horns.

The journey through game art continues.

Week 7

Sentry Gun Overview

This was a fairly simple project, the hardest part was taking all my melting pot ideas into one gun, I feel it worked out well in the end. The first part was research and I picked pastel goth and diesel punk as my themes. To make the work I took mechanical but rounded shapes, made a bash kit and started working. The outcome was suitably diesel fuelled and so I made the colours, details and feel for the turret more pastel goth. The whole thing turned out a lot better than I thought it would.
Next I moved on to modelling, this wasn’t hard but not as difficult as I had thought the blimp like back end would be ethier so all good there. The unwrap could have been better, I didn’t know how I was going to recreate my ideas so there are parts I would change. Eg: the separated ID’s could have left more space for my other parts and got more space themselves, however as is they could share the same roughness maps.

The poster i made for the graphic side of project.

Engine was a lot of fun and experimenting, especially with textures and my new toy the Lerp node.

I’m really pleased with most the topology on this project, I tried to get it to flow well and use my large 5000 tri budget efficiently all across the model. Some of the wires look similar and would share texture space, but I didn’t want smearing so I had to lay the over each other rather than reassign verts as I normally would.

My projectile was an eye ball, I’m really happy with how the normal map pumps up the veins, and it’s also really well fitted for my gun.

Using blue Prints to get the double barrel working.



Learning about lerps, I love lerps, I used several on the glass to give it hue and opacity (glass material above). Lerps are also what gives the gun a purple pearlesance and that’s perfect for the look of my turret. I feel it looks less toy like because of the lerps.

I also did a little paint over to show what I would change, in this case I would make the purple glow out the back reflect off the metal more.

I’ve really love learning about texture map and how I can use I’Ds in max to swap out and become engine textures, which are not part of my texture budget, so that’s really useful. I also love the way that Lerps work and I may have gone overboard with them. However I think if I went back once again my concepts weren't up to scratch and I should have shown my progression more. I got some feedback that my pivot was too thin (oddly balanced I think they meant) but didn’t have time to re unwrap and change that so that was a shame. I looked at my idea and changed parts and did a lot of working out with how I would create this gun in 3D, but I seem to not have shown these processes anywhere, nor have I shown how I went back into the albedo and added more details and scratches and decals. Which is a shame because it was a part of how this developed. Overall I’m happy with my result and it’s a project that I would like to revisit and add more too, eg: animated eyeballs flowing through the tubes like they were meant to, but I couldn't work out in time.

In engine shots of my pastel punk diesel gun

Week 6

Back on the Grind Stone

Modelling is love modelling is life

3DS Max is the main tool in my software belt as a 3D modeller and my relationship with this software has strengthened as the course has gone on. But none could have anticipated the struggles of low poly bat wings.

The Model

Nevertheless I have prevailed and me and Max have created my turret and it looks just like the silhouette, complete with cage pivot. I'm really happy with how the pivot translated into 3D.

As it was modelled I felt it was time to test my gun in Unreal 4; in hindsight I should have had it unwrapped but I was too excited. This showed me that my favourite part, the caged pivot, looks odd as a pivot when the gun moves. (Not to mention the 30 times I tried importing the mesh only to have the gun parts move in 3 different directions.)

My Topology

My attempts to make it look more suitable have failed, but I kinda like it looking off. The gun itself is a little weird and wonderful and maybe the insane pivot is just an aspect of that.

The best part of all this is the blueprint learning that has occurred through trial and error. It may not be an opinion shared but trial and error makes me clever. Every project I have a struggle with I learn and can apply so much more for my next assignment. 

Week 5

Individual Projects Begin

Sentry Guns, Iteration for the Nation 

A new project begins, the sentry turret. This time our brief is to design, model and transfer to engine a sentry gun. The blueprints and animation is set up. All we need to do is import and replace... or so we thought. (Freddy and the pivots from hell is another story).

We began by making a good old fashioned pinboard: http://uk.pinterest.com/cantondmuga/gunned-down/

Oh Pintrest the way you make my research easy and accessible makes my heart skip. My chosen theme was pastel goth and diesel punk. My idea was to create a diesel punk silhouette, then theme and mix in the pastel elements. 

From here we created reference boards and from these our first silhouettes. My first silhouettes were made from a bash kit created from my mood board, which was really good fun and a technique I hadn't used before. I could have iterated until the cows came home nothing was stopping me. However when I got to 50 reason stopped me, it was time for feedback

From the favorite designs of each person I asked, I produced tonal outcomes and started to develop them into something more pastel goth. Needless to say, not much time was spent on this after all those first silhouettes. At least i'm happy with the sheet and feel a lean towards which is the strongest design. Before I knew it we were knee deep in the turret project and I was on to colour sheets.

Tonal Studies, how will these be 3D

Colour Sheet

Final Design, and I'm very excited 

Next time on Freddy does game art, 3DS max and the day of the round shapes.

Week 4

The Film Room Post Mortem part 2

Once we got the ok, it was time to go into overdrive, turn it up to 11 and burn some midnight oil. Although if I was to redo this project I would have started once we had a film chosen and probably taken advice from tutors rather than relied on the dragons den method we were given, as we spent far too long deciding and not enough time working.

But then you never know what issues a project will throw at you, learning is a mixture of failing and picking up the usable pieces for next time and successes where you find a working method and roll with it.

Organisation would be key in this project, something we didn’t seem to realise until week 2. However once it was introduced we used many methods of keeping track of each other, our work and the group internal deadlines.

We used methods like: mood boarding the scene for assets, colour pallets, the waterfall timetable and an editable schedule to help us all add in parts that we missed and crack the whip on each other.

Now it was time to concept for my assets. A lot of them were front on or difficult to see in the view of the tent we chose. So for the trophies, the part I couldn’t see as well, I did the most iteration. In the end I used line orthographics to obtain the shape I wanted. I think my concepts are poor and if I had gone back I would have started them earlier so I had more time on them.

orthographics

Because of my concepts being poor, I decided to use the block mesh we had made, to create simple, visual, orthographics. These were used for presenting and helping understand the paper models proportions, without having too much topology in the way of the asset you are trying to measure.

From concept to model, the time to use 3DS Max had come. And I was ready. Although I’m happy with my models, there are parts looking back that I would change to fit the tent better that are a little off in engine. The candle holder for instance is correct in max, but in engine the camera used stretches the model too wide. Also because of my working view the globes holder is both too wide and the turner is too far down the holder.

Once I had unwrapped and textured it was a little too late to fix this because it was lit and set up in engine ready for presentation. There will always be an argument for if I had a bit more time. However I really feel that if I had of used the time I had differently, spent more time on concepting and getting feedback, then these where issues that could have been ironed out of our scene.

Models in engine

using masks for emissives

Complicated textures

Texturing was reasonably simple, but I was using a new map, an emissive map, which is a map that masks out the non-glowing parts of the texture so only the correct parts glow. This was simple but plugging it in, getting it to look right was another matter.

Introducing Freddy to complicated texture blueprints in engine was also not a simple task, but we got through and I now understand a lot more about PBR.

Least we forget the paper model, a hallowed construct of paper and love. This was something that we put a lot of effort into, although not much time and it somehow came out the other side looking pretty good.

I love it because it’s cute and it was a lot of fun to do together. However It was really successful too, it matches the scene well and we faced similar camera issues with the physical model as we did with the engine one. Because of the focal point of the globe, the lens wants to focus on it or the back, and not both. 

To create the photo of our model I took it on my DSLR, so I could fiddle around with differential depth and then composited two successful shots into one.

Overall I’m chuffed about how well we worked as a group, I’ve learned an incredible amount about engine, group work and most importantly I’ve taken on all these mistakes and learnt from them. I may have changed parts with time, I may have not produced my best work ever but I learned so much and I’m at uni to learn.

There is no failure in mistakes, failure is not learning from them.