Tuesday, 28 October 2014

Baking... yet another eternal quest

Baking my Rocket Raccoon Railgun down has been annoying to say the least but I may have got to good point and definitely have learnt a lot about baking in the process.

To start with I was talking to Rich about baking and he imparted some of the wisdom from his own quest for the perfect bake. As I this was a hard surface project I took Rich's advice by setting the smoothing groups up so that anything with a harsh angle was set to be a hard edge on the low poly and then separated the UVs so that the different smoothing groups weren't welded together on the UV. It seems that most things talking about baking explain this as the way to do things including Scott Homer's presentation that was everywhere last week if you're interested in this stuff this presentation is awesome to study! Go study!

This technique however didn't lead to the desired result, I can't say why... computers are evil masterminds of rationality so I must have done something not quite right. You can see that the edges are unrealistically sharp and in comparison to the high poly the baked version just doesn't look right. So I continued to play around, got fed up and set everything to one smoothing group as I had done before and that ended with a blobby mess.

Then I realised that the normals were being forced to, in effect, work too hard. The shape of the mesh was such that having everything on one smoothing group was forcing the normal map to counter the effect of the smoothing groups and straighten out the surface again. As most of the pieces in the asset have inards and are kind of like 3D "C" shapes I realised that having the same smoothing group going round the outside and inside was just a bit too much so I applied a different smoothing group the inside of the "C" and baked it down, here is what the edges look like in comparison to the high poly.

Ignore that smoothing error on the right side of the low poly I'm working on that.
During these series of experimentations I modified the shape of the low poly to fit better with the sculpted high poly, I took off corners that were too harsh before etc. I'm now in the process of going through everything and re-baking and adjusting the low poly I have a feeling that this should be worth the time I've spent on it. Unfortunately the normal maps themselves look like they've got worse but their effect has got better...

Blogger's compression has got worse :(

Put simply I'm confused :) This is not a tutorial on what to do but more my experience so far trying what should technically have worked and didn't...

Monday, 27 October 2014

Less Than a month left... 0.0

Before I started at Ubisoft I was really focusing on my understanding of presentation and composition. I was confident about my ability to model, texture and use the engine I was however struggling a lot with composition and presentation. Fortunately the work I've been doing here has allowed me to continue practising those things. At points I have felt like I wasn't learning anything but have realised that I haven't been learning exact techniques or something that I can say "This is what I have learnt" instead I feel I have improved the more difficult to define... stuff.

The range of work I've been doing is quite large, from the almost daily need for text to be changed on a template to the continuous endeavour of finding the right places to show off the DLC but the over-arching work has pretty much been composition and presentation. This is a really long and not very exciting or engaging process as knowing the places to go in the town to find worthy compositions and really nicely refined assets is difficult and much of what I need to do is roam around the gameworld. The open world nature of the Assassins Creed series means that finding that perfect place to take a stunning screenshot is actually more difficult that in a more linear game. I'm finding that I need to repeatedly try and imagine I've been given a pretty buggy camera and being told, go and find the most impressive way to show off the town. I'm normally really good at taking far too many photos and find a few gems but this feels so different.

Today came one of the major things that I have had to confront about working in the industry. The work that I've been doing for the past few weeks was on a rough brief that I thought had been decided upon by the people in charge of marketing ACU: Dead Kings. It turns out that I have spent the past few weeks working on a very rough brief that was just a first draft brief and hadn't been talked about in much detail. The meeting was great in terms of ironing out all the problem areas with the proposed screenshots but it also meant that I have to ignore the work that I've been doing and go back to the early research stage of the process, essentially losing a couple of weeks of work.

Spent about a week making a more colouful cover for M!Games
They went with our original proposal of the game's Key Art

Looking back I can remember a huge number of times I have spent days or weeks working on something only to be told that they want something else; for technical, artistic or even third party reasons. This is something that YOU HAVE TO GET USED TO. Quick turnarounds are essential as it doesn't matter how much time you have spent on something if that thing is not needed... its not needed. Start work straight away on the new brief but remember to keep as much work as possible just in case that original thing is actually needed in the end. Briefs aren't clearly defined and often the people setting the briefs haven't yet had the time to develop the briefs or they simply haven't talked to other people about the briefs and so they are always changing, flipfloping and in general they aren't definite.

Such was Uni, Such is Industry. Such is Life

Friday, 17 October 2014

FMP Idea Dev: NeoMediterranea

As a long time fan of Arena Shooters, as I'm sure is understood from earlier posts, and this summer has been pretty exciting. Unreal Engine 4 was announced and released in March. A few months later after two years of Epic saying they aren't going to make a 4th Unreal Tournament, they announced that they would indeed be making a 4th and this would be made with immense help from the community. Following this Bosskey studios, Cliffy B's Indie team, made some noise about another free-to-play title but nothing more than that. Following that Reakktor Studios, behind the Necron Games, announced that they were bakk with an answer to the current state of the FPS genre, Toxikk. Toxikk is by far the closest to the end of production and it plans to answer a lot of problems that have arisen with the yearly model of the AAA FPS games.

This renaissance and people at work talking about how little the artists thought about gameplay lead to this idea.

"To create a competitive dual map for an arena shooter using a modular asset pack usable for a series of maps set in a similar location."

The idea is to produce a pack of modular assets, ranging from materials to fully animated prefabs containing multiple meshes. Where possible these assets should have exposed variables allowing the user to change as many different aspects of the asset as possible for example: material, colour, age/grunge, speed of animation etc. The assets must all fit within the art-style and be as modular and reusable as possible. I have been thinking about the architecture around Montpellier as being the main inspiration for the environment as I feel that it is both interesting and fitting for the environments for arena shooters. The Mediterranean-Way is to fix things so that they do the job, if you look at Venice this is obvious, you have ancient buildings that are repaired and repaired and rarely demolished and rebuilt. The same is true here in Montpellier on the other hand it also has a young population and it seems quite successful in modern industries therefore there is much modern architecture and even some very modern features around the city. This idea of the contrast between the old buildings that are being patched up and fixed and the brand new modern or even futuristic buildings being right next to each other is really interesting. I want to push this idea even further into the future hence NeoMediterranea. I'm going to need to get some more photos of the modern architecture in Montpellier but here is a moodboard of a selection of images out of the hundreds that I've been using as reference.

After having this idea I realised that my level design hasn't been tested in a long time. As I've been playing and watching as much Quake Live as possible and I'm very keen to combine art and sport together I decided to make a rough blockout of a Dual Map. Dual maps are really important to the competitive scene and the level design is very tough to get right because of the level of skill of players. Arena shooter duals aren't about shooting they are about map/item control and therefore the level design and item placement is incredibly important. This was tough as I haven't really done any level design based things since 2010-11 so I had to re-make lots of mistakes and remind myself of a lot of key points but here we are DL-NeoMed_MK1. 

So far I've started to photobash and concept some of the areas. I want 3 different areas so that a range of environments are plausible with the asset pack, old, broken and abandoned area, futuristic, Tokyo/modern inspired area and industrial sewers area.

Sunday, 5 October 2014

Rocket Racoon's Railgun

Baked down with some VERY early and basic materials applied thanks to quixel. Definately going to be working on the metal material its really horrible at the moment. I am looking forward to making the funky blue glass/metal/stuff and the effects.

Just a reminder I'm working from Maciej Kuciara's concept for Rocket Racoon's gun.

Maciej Kuciara

I Did Stuff!

Here is some of my work, in-game photos of Assassins Creed Unity: Dead Kings. This job has definitely been more about photography than anything else.


I may talk about the process at some point but its not particularly interesting a lot of thinking about composition, a lot of talking to art director and getting paintovers...

Past Halfway - Stop Thinking of Cool Stuff and Do Cool Stuff

I managed to miss all of September (started writing this in the first week of September). I have recently come to a conclusion about personal projects that I was told over and over, I thought I was listening to the advice but now I don't think I was. It was very easy to convince myself of possibilities when in student life which even if incredibly busy has a feeling of "there's time".

One of the main things I'm always worrying about is the future; getting as good as I can get, improvement and not seeing the desired improvement and recently missing a term of Uni. My brain tells me that the 2.5 month term would allow me to play with more cool and different things so that I could fulfill my desire to be a T shaped artist being fairly strong at a lot of things with one real strong side. In reality the term would likely give me an extra environment for my portfolio and probably allow me some time to improve my FMP ideas, which I hope to write up into a blog while I work them into presentations.

Life will happen and I will end up somewhere, I know I am dedicated and will strive for what I want. I just need to strive positively... have an adventure!

Talking about work after work, onto personal projects... I know that I am an ambitious person and I am an excitable person. My somewhat naive approach to personal projects seems to have become more and more clear with the worry of losing a term. Because of this I have tried to shrug that concept that "there's time for that". I don't really want to think about all the projects, or ideas as I'm now going to refer to them, that I've come up with and have fleshed out in my brain and talked about with people and got excited about and then just simply not followed through on. However here's a nice list: Quivering Quacks (stylised pub started at the end of first year and looked at a bit in second year), Dockyard (BMW art test environment), Breeding Pool (small environment based on a painting) and Caligula's Palace. Almost all of these have been started but not got vaguely close to realisation, the reason seems now to be an over-reaching idea, too ambitious and too much of an "idea" rather than a project.

I have been told multiple times that an entire environment is really difficult for a personal project and maybe to focus on a small area and do it really well, I thought "well I know how to do it and I reckon I can speed some processes up so I can do it". The other problem I've found is that because I'm still learning new techniques and processes and I'm trying to do things to the best of my ability things take a lot longer. On top of this there is a lot to organise to create an environment, all the different assets and different elements, it can be rather overwhelming when you can only spend maybe a couple of hours every couple of days. After getting more and more frustrated about how poor the Caligula's Palace environment was progressing and after talking to some "5th years" about personal projects I decided to take a bit of a break from Caligula's Palace.

Instead I spent September working on one asset which I will try and make as highly defined as possible, I will really focus on the material definition, high poly modelling and hardsurface detail techniques but mainly just have one project to focus on so that when I get home from work or have a couple of minutes at work I can sit down and know exactly where I am and just continue. I'm also planning on doing this as its a smaller project which will show results quicker and also teach me things, too often have I learnt a lot without anything to show for it. Hopefully I'll return to Caligula's Palace with slightly less expectation, different techniques and a new desire to make something complete.