Smuggler's Cove Post Mortem

Name: Chris Lonsdale

P Number: P12206538

DMU Email: p12206538@myemail.dmu.ac.uk

Email: ChrisTheJam@gmail.com 

Portfolio: http://riversofredness.wix.com/chrislonsdale

Name: Lewis Gibbons

P Number: P13232166

DMU Email: p13232166@myemail.dmu.ac.uk

Email: lewisgibbons12@gmail.com

Portfolio: https://www.artstation.com/artist/lewisgibbons

Project Overview: We aimed to create an environment based around the theme of pirates and smugglers. Visually, we wanted to execute a heavily stylised, hand painted look. Our original inspiration drawn from Riot's League of Legends and Blizzard Entertainment's World of Warcraft. This is a very high bar to hit obviously, but we used these existing worlds not to belittle our own work, but to have a constant point of solid reference. Whether you like these games or not, they work and both of the companies behind them are extremely successful in what they create. 

We wanted the scale of the level to be quite vast and expansive, focussing on the vertical elements primarily, and always having the feeling of things being a lot taller than you and towering over the things beneath them. Whilst trying to give off a sense of large scale through big buildings and inaccessible terrain, we also needed to pull off a level of intimacy when walking exploring the various parts of the level. Striking a balance between broad strokes and finer details was sometimes a challenge, and the amount of hours required to populate so many different areas with various assets and keep the composition good, and to make sure things make sense. 

Incorporating a mixture of hero assets that added character to the level and modular pieces that could be used to flesh out such a large level was another balance we had to be constantly mindful of. On one hand, there simply wasn't enough man power and time to make every building and repeatable asset completely unique, however we needed to have enough separation and small detail between each area that gave good readability and looked believable still. Whilst there's always room for improvement, I think we took a solid approach to the creation of the level. Starting with mood firstly, and establishing a couple of solid modular assets followed by a few smaller ones; we had a good base established in the first week of the project. 

There was quite an interesting correlation throughout, I feel as though the rate at which assets were produced at the beginning was very quick, but then as the project progressed, our standard raised by quite a significant amount. By the end of the project, namely from May until now, the rate of production increased again, but we were at the peak of quality for the project; so producing our best work but at the speed we were when FMP started. This gives us good momentum to carry on, in the build up to particularly our degree show in a few weeks, there is definitely the time and opportunity to keep taking old creations and bringing them up to standard.

There were some significant changes in our level as the project progressed that certainly need to be addressed. The first is the change in direction of style. Whilst subtle to some, I feel blending a (sometimes) PBR approach with purely stylised hand painted work created a successful outcome. More importantly using Z Brush and Substance Painter in our workflow generated quick and solid turnout of assets and environment pieces alike, ranging from sculpted tiling textures to whole unique buildings. This change in workflow encouraged a change in reference of style somewhat. With the much awaited Overwatch becoming more public and recently playable from Blizzard, we both noticed how our visual interests were drawn heavily to it, and it was something we tried to replicate quite a lot in the last few weeks. 

The second was quite a dramatic change in mood. If you go to the earlier posts of this blog, you'll see what looks like a completely different level. Whilst I earlier mentioned that we set out to establish the mood early and as soon as possible, that didn't mean it'd be final. It was nearly three quarters into the project that two industry visitors from Creative Assembly came in and had a look at our work, with the mood and colour scheme of the post process (green fog) being out of place, in their opinion. We agreed, and together looked to find something more fitting. Going from a dark, moody scene to what we have in the end result is quite a shift I'll agree, but a decision we both agree has a better visual outcome.

Beauty Shots: A selection of screenshots from in game are posted below, with a couple of them above as well to break to the walls of text! 

In the last screenshot above, you can see an aerial view of a good chunk of the level, to give some indication of sizing. Below we've posted some asset renders that we've both made, to try and show a breakdown of a handful of the assets we made for the level. Doing a render sheet for every single one would take many more hours than it's worth, but here's a selection of the work we're proud of:

Conclusion: It's been very hard to summarise twenty weeks of continuous work, as there's been a huge contrast of menial tasks feeling like they'll never end, and monumental design decisions being made in a matter of minutes. I think the most successful element has been the learning curve that has been achieved, and within that the ability to learn as a unit of two rather than separate individuals working from home alone doing our own thing. Coherency and communication are two words that are key, and have remained key through out the whole time we spent working on this project. Initially, we were never really sure where this project would end up. We knew we wanted to create something quite vast, and the aesthetic of it to be something colourful and stylistic; two things I think we've achieved well. To reiterate on a previous point however, there is almost some frustration mixed in with the satisfaction of getting better at what we want to do. What I mean by this is, whilst consistency has been kept between what the two of us produce, the standard has risen by such an amount, that, as the creators of this project; it's damned easy for us to spot the weak areas. 

No doubt we can go back and improve these parts, there is still time, whilst writing this we have just under three weeks until we present it professionally at degree show. It's rather cliché to say, but it's been a roller coaster of a project, there have been times where motivation levels have gone into negative, and the thought of painting another texture felt critical to our health. However, we can confidently say we thought out an idea, stuck to that idea, and executed something that we're pretty happy with in the end. The pair of us have an expansive list of things we'd like to change, and critiques to act upon, however feeling confident in having a pretty solid foundation to start with offers optimism to produce something even better now. With a couple of very pivotal decisions happening throughout the project, it's been great for our sanities to have the other person there. Working as a pair has consciously raised the standard; you always wish to bring your level of work up to that of the other persons. With critique and feedback being so regular, we, for the most part, safely avoided the delusion of thinking our work is always satisfactory or great, and as a result of this, naturally got a lot better. People may say time management is more critical when alone, but constantly having the pressure of another person (or whole team in some people's cases), always weighed down on you to the point that you kick your own butt into gear more often than not, and productivity is on a good high. 

It's been great to utilise different workflows and software during this time. Thinking we'd originally be sticking to the bread and butter of 3D art (3Ds Max and Photoshop), it was actually way on the opposite spectrum. Using Substance Painter, Z Brush, Marvellous Designer, 3D Coat to name a few, we really sped things up alongside being able to get a deeper knowledge of them all. 

It was previously mentioned that we both have quite a long list of things to act upon to improve this level, however it's important that some of the more key points are highlighted in this post. Reflectively speaking, if we had stuck to the traditional hand painted, low poly, diffuse only approach the whole way through, I don't think we would have been nearly as happy with the final result. All the more, we wouldn't know half of the things we did now, and with regards to the increase in production rate that things like Substance Painter has granted us, I also don't think we would have made half of things we did in the end either. So the transition to a new style and the blending of an old one is quite a risk in some people's eyes, I think it paid off though. However, I have to say, that I'm really quite curious of how it could have looked if we followed the latter style the whole way through. I (Lewis) wish that I had learnt these new software packages sooner, and taken the time to become comfortable with different workflows before diving into a project as big as this with someone as knowledgeable as Chris. It's important to highlight that, without his help and our seemingly identical brains I would have found this project a lot more difficult than it was.

Aside from this, our general direction is going to be following the logic of 'if it doesn't look good, polish it until it does'. Quite a key point, We've chatted on about speed and producing assets quickly, but haven't talked quite as much about the quality. Now that there is no looming deadline, we can invest time in to things and keep pushing the standard as high as possible; and more so bringing the things that slack up to the level of everything else. 

Overall we're really pleased with what we've produced, and this has been a great end to a pretty intensive degree, it's exciting to see what we can do with this project and projects to come!

   

Cove Completed - Week 20

So this is the last blog post of this project and degree! Well, minus our post mortem blogs, which we'll both have posted later today. A ton has been learnt through out this project, and we're really pleased with the final result, and feeling good momentum to move on do more cool things. I'll start with some renders of the level that we handed in today. Things may change in the build up to degree show! There's more to work on for sure, we'll aim to polish as much as we can for the big day.

So that's the project drawn to a close for now. We're trying to work out how to get a high quality video posted online, as places like Facebook instantly compress the file and quite a lot of the quality is sacrificed. We've really enjoyed working on this, and in future I won't underestimate how hard it is to get things looking with a hand painted approach.

It's incredibly weird looking through this blog and seeing a visual documentation of the last 20 weeks, our project has changed so much it's weird. I also feel a lot more confident in the work I'm producing, things really started to click towards the end, and I'm sure Chris can say the same. There's a very noticeable difference between things created at the start and then things more recently, from small assets through to to whole buildings. In our post mortems we'll be analysing the project as a whole, giving more of a breakdown of everything we both did for the level and talking about what can be improved both with our workflows and the final result itself. Can't believe Uni has finished as of this point, we hope you like our level! It'll hopefully look even more refined for the professionals to come and see at degree show.