Sunday 25 January 2015

DMUGA Week 17 - Texturing the level + BMW Interview

As the project continues, we finally get round to texturing the majority of the assets; making sure all of the models are done correctly, I'm in a rush to get everything together for Thursday's interview, and there's just so much more to do!

First things first, I found out on the previous Friday that I was selected for the BMW interview, so, over the weekend/ Monday evening I was adjusting my CV and beginning to create my cover letter in preparation for Thursday.

There was a lot of problems on Monday with the lighting that we tried to fix. The issue has been there since the start, but I thought it would have been solved after we did the lightmaps for the containers. This wasn't the case, so we had to sprawl through the internet to try and find anything we could that affected how the lighting reacted in the scene. A few settings were tweaked and it looked ever so slightly better, but it was still apparent!
Just for reference, this is what I'm talking about:

Light bleed still there... Trying to find a way to cover it up in most places.
In short, 80% of Monday was just trying to solve the issue, or come up with ideas on how we could hide it. At first, I felt like I had wasted a days worth of work, but on a personal level, I found out some new things, and did come up with a way we could just hide a lot of it from the player.
The last 20% was on making a couple of assets, and trying to get some sound effects in to make it a bit more immersive.

After this week's life drawing, I had a realisation about my work in that I don't let it flow as much as I would like it too. Also, I'm not going out of my comfort zone to improve as much as I'd like either - Things have got to change. What I need to do:
-Try different media
-Try a different approach e.g. focus on one area instead of the entire body to create a better composition.
-Relax more, take a structured view i.e. plan the key points first so it doesn't look distorted.



In the afternoon, there was another issue that needed to be solved... The frame rate as you are walking through the starting area drops dramatically because of the particle emitters. There were too many particles at once, so I got a tip on how to create LOD's for them, plus, reduced the lifetime of them so that it wasn't spawning and hanging around for longer than it really needed.

Add LOD
Duplicate the settings you want to change from the higher LOD
Change LOD Settings + Emitter Settings to what you want
There was panic in trying to get some images online for my portfolio that I had to get ready for Thursday's interview, I would give you a link... But it's definitely not something I would want to be of much consideration in the current state it's in. Not to mention, I hadn't even got anything smart to wear as I assumed I remembered to bring up some smart clothes - not the case at all! I had a tie and a black shirt, but no shoes or trousers... Off to ASDA at 9 o' clock on Wednesday evening!

I also spent some time creating some seamless textures so we can use them over and over if possible:

-Concrete
-Burlap
-Tarpaulin
-Sand_003
-Rock

By Wednesday evening, I had my CV, cover letter, Portfolio and blog link e-mailed, so all I had to do was go to the library to print them off and prepare.
Work-wise, Wednesday was a good day as we got some more texturing done, the majority of the assets were coming together as each person is doing their portion of unwrapping & texturing.

After waking up on Thursday morning, it became more and more daunting the closer it got to my interview time at 11:30. We had a Q & A session on internships until 10 o' clock, so I thought I could get some work done before I went in, but I was so nervous that I couldn't focus. In short, I ended up going in at around 12:30, and not getting much done at all.
Unfortunately, I didn't get to the last 2, but I enjoyed the experience of being in that kind of situation as I hadn't been before, not to mention I actually was forced to tune-up my CV as it wasn't very good before, AND, I wrote my first cover letter. So, I know it sounds bad I haven't learnt all this already, I know now how to go about this kind of process and what to expect. Win/Win!

In terms of the project, the whole morning was on the interview, so I only really had the afternoon/evening to do anything worth-while. I did add in some more sound effects, change the lighting slightly so it's a bit more natural, and finished unwrapping/lightmapping the rest of my assets ready for Friday, as we agreed to get everything together and imported into Unreal Engine 4, so we can start populating it.



Friday was a little bit hectic in terms of managing everyone's files and putting them all together into one because we didn't want their to be any problems, but it turned out some models had problems so we couldn't export, the file re-naming & re-importing was procedural but still required us not to over-write any of our previous work etc. To summerise... I had a big headache by the end of the day, but everything was re-imported into the engine successfully, so now we can just focus on finishing the textures, populating, and re-fining the level in more detail.

Here's this week's run-through of the level so far:












Sunday 18 January 2015

DMUGA Week 16 - So many assets... + Blueprints amaze me!

This week we have lots in store including making blueprints, beginning to texture, a presentation, and some extra fun parts! This includes my mediocre attempt at a sandstorm particle effect!

As the title suggests, I got to look into how to do some blueprints in UE4 in a bit more depth than I have done before. Last week I got the ladder working well, but there were a few more things we wanted to include which are quite simple and easy to do e.g. spawning after death, sprinting & crouching. In order to do these I got help from Chris - A student on my course. I'm slowly getting into the swing of blueprints and what you need in order to do certain actions, but it still confuses me a little.

Crouch, Sprint & Torch
Container Blueprints
We have a lot of different blueprints of the containers because it means we can easily switch between the ones we want in a couple of seconds instead of having to re-build them in pieces, plus, we can make individually unique containers by having some with doors open at different angles and with different side panels/doors. - EDIT: This is not the best way to do this as it's pretty expensive - just stick with separate static meshes and groupings!!!

I also did some more modelling/unwrapping (including light maps) for the rest of the day.

Tuesday's Life drawing was generally ok, although I didn't use a fine-liner or watercolours like I was going to. I'm weakest when it comes to colour, so I wanted to slowly start to incorporate colours into my drawings. Next week I'll definitely do fine-liner for the quick 2min-15min poses, and watercolours for the longest one.
That doesn't mean we weren't posed with a challenge, as this time the model had draping fabric which was very shiny but see-through so it was quite hard to interpret it.




We also got some helpful feedback from the tutors in general, and some advice on how to separate out the texture budget as this is something which can be a hard thing to do when there is so many assets already made and that need to be made - They need to be divided onto the sheets and assigned so that we know what is possible to do with the 10 1024x1024 sheets.

I managed to get the gate modelled, re-arranged some parts of the layout again, and the textures made for the two types of wire meshes + barbed wire.




In addition, I made some modular pipe pieces so we can make any system we want which still matches with the dimensions of the containers:


Wednesday was a good day because not only did we get a lot of work done, we had a visit from the CEO of Skylands - A voxel based game which is like a vamped up minecraft mixed with Zelda/other games to create a stylised, adventure game which is procedurally generated to keep it interesting all the time. Not going to lie, the presentation made we want to try it out!

Anyway, I got down to modelling some more assets like the weapons we wanted to include from the film "Shaun of the Dead". We wanted to include these in there as an easter egg just for a bit of fun for the player:

We also have a cricket bat - will group them together in the level with a note for Shaun next to them
I also made some separate panels with cuts in them so we can use them to vary the assets of the game engine. By doing this it allowed us to properly map out the player's route and walk through most of the level so we could get a feel for what it is like to run through it. In addition to this, I felt like we needed some audio of strong wind and some kind of deep rumbling when the player reached the pit, just to add to the experience some more.










The only thing which needs to be done before we can get through the whole level is the blueprint for the gate/switch box, but Andy was already cracking down on that and trying to find out how to get it working.
I also looked into how to get a night sky as we thought it would be nice to have harsh lighting of some floodlights beaming down into the level. However, I couldn't get it looking how we wanted to, so the next day, I looked into getting a sandstorm effect as opposed to a night setting.

Presentation feedback = Was a lot better than I expected, the only trouble is the scope of the project and whether or not we'd have enough time to finish it and get it looking nice. I think we can do it, but we really do need to get the texturing done so we can have half a week for polishing & populating the world.

I spent some time creating the dust particles and 'cloud' particles, trying to get the right look for them both to make it seem like you are travelling through a sandstorm. I have to admit, it looks pretty nice, but I want to add more to it e.g. some whirling & whispy swooping action - it's ok in the mean time though.

I also got a couple of different fire particle effects working. I produced two because the first one I made wasn't that great because you could see the panning texture, so I was ok but not too happy with the result. I found a second tutorial on how to make fire, and it looked great! It was around 3 hours of work, and it took me a lot longer due to some audio lag, but the tutorial is really good and I highly suggest anyone to use it (For both UDK AND UE4 as 98% of it is transferrable!):

1st Attempt
2nd Attempt - Imbue FX Tutorial
Fire in unreal engine - Imbue FX

Speaking of tutorials, I also had a look into vertex painting because I have been meaning to get into it for so long because of how good it can be, but not got around to doing it, but I feel like I desperately need it now due to the variation in textures that we require. So here is the tutorial I followed and it's very basic, but very helpful too:

Vertex Painting Tutorial

To round it all off, I did 3 bits of tarpaulin so we can use them in the living area of our map, and some texturing work.
So overall, a good week for work as we got a lot done, but there is still so much to do and time is running out... Not to mention a very large problem we still have from the first week which is the light bleeding through into the containers:


Saturday 10 January 2015

DMUGA Week 15 - Back from Christmas break - now on the way to container city?!

Back from the Christmas break and we're straight into another group project: Container City. Over the holidays I did take some time to relax because you just have to - It's family time. I did however, put together one HUGE blog post giving you a breakdown on how my summer project went and what it entailed. Found here: Summer Project - Ancient Ruins

Anyway... getting back to Container City, we have to get into groups of 4 and design a part of a level using modular assets with interactive elements based from 1 of 2 themes: Sci-fi futuristic, or dystopian post-apocalyptic. Here's how the first week went!...

Our group consists of: Andy, Georgia, Luke and myself.
The very first thing we did was decide on which theme we wanted to go for. We didn't mind as we had ideas for both, but were inching more towards dystopian because it would give us a chance with the challenge of modelling and texture lots of assets which need to look less clean & perfect. It became apparent that you could combine the two and go for sci-fi grunge, but we were 90% sure on this one (I did create a sci-fi moodboard just in case however).

There were a couple of ideas being flown around for the setting, why is this here, and what would make sense it include, but a key comment in the briefing was to focus on a block-out and how to make the player route an interesting one before settling on the context too much. We put together some block-outs in 3DS Max first just to see what we liked from each other's routes and this is what we got:


... A big mess... But it got us thinking on what kinds of things we want to include to make the route fun for the player. Andy had some really cool sketches for his thoughts on the route and how we could vary the lighting conditions so that it is dramatic and interesting - A key part of the brief.

I put together a couple of moodboards just to get some ideas going on the types of things we need to include:



Tuesday was the day we began to look at the player's route in more detail and what we could do more specifically to enhance the route. The brief states that we need interactive elements, so we thought why not make the player have to go through a small challenge to get a key, then use it to open the main area (of which we didn't know what it was going to be). Here's what we were thinking:



Our initial idea revolved around something which we were yet sure of... the context. It involved climbing up a ladder (interactive element) from the sewers (dark place), into another dark area such as prison cells, and finding a key somewhere in an outside courtyard, going back into a gloomy tunnel with volumetric lighting, completing a very simple jumping puzzle, finding another key, and then that will unlock the big area.

As my description may suggest, this route was far too long in the period of time we have to make this. So, we cut some bits out:



Before I carry on, the life drawing session this week was really good as it got me back to drawing, but my outcomes weren't something I'm proud of:



It's not just a bad photo either
It may just be the first L.D back, but I know I can do better. I really wanted to go to the life drawing in the afternoon as well, but the container project had my full attention at this point and I didn't want to lose any time on it.

Moving on, we changed the player's starting point and had them finding a way to open the gate, and then sliding down a long container into the courtyard. We thought this would be a really cool thing to include to make it fun. However, the a 3D tutor advised that we'd probably have 1 person making the rocks at the start looking good for like 2 weeks so it wouldn't be that viable.
We kept it in for 1 day but then decided to take it out and think of something else to fill in the starting point. Nevertheless, here's the work I did on this route:





We were also advised to cut the level right down because it was still far too big, and this is what we ended up with:


Two big blocks standing up are watchtowers guarding the city against whatever it is they want to keep out... Zombies?
I thought it was ok having to change the route again as it made our lives so much easier for scale-wise, but I had a change of heart because I realised that now, the route could be completed by the player in about 20 seconds, and it didn't include anything that was really that interesting.

We tried to get it in balance on Wednesday/Thursday so that it wasn't too big and wasn't too small. This is the final layout that at the moment is roughly what we're going for:

Current layout
Current progress with actual assets
Misc Assets
We've spent the rest of our time this week getting a lot of the assets made because there is just so much we need to clutter the map with. It just worries me a little that we only get 10 1024x1024 texture sheets/40 5125x12, but we need to have many different texture sheets for each object to get the variation in there.

In addition, I got the ladder blueprint to work with my meshes in the game engine, so at least we know we can definitely have that in there now as an interactive element!