Friday 25 December 2015

Merry Christmas

Hi to all my readers,

I just wanted to say Merry Christmas and thanks so much for coming back and reading my blog. I hope that you will continue to come back and keep reading about what I am up to in 3D and computer games. I wish you all a great day and that you have a great day spending time with family.

I have to sit down between now and when I start at Playground games and learn a few new things. I will keep you posted as I plod along like normal.

Until then enjoy your Christmas where ever you are and I will see you back here soon.

Friday 18 December 2015

ZBrush and Art test Update.

Hi all again and welcome back!

I thought I would give you a bit more from my progress and where I am with everything. As you may know I have been busy brushing up on my ZBrush skills with some simple speed sculpts and trying to learn all the new features that have been put into the program since I have last used it.

To be completely honest I have really skirted round using the program because I didn't understand how it places new tools in scene so I am just forcing myself to get used to it. I have done a few 10 min speed sculpts just to get back into the tool sets and getting different forms. Rocks in particular I am working trying to get organic wear and tear quickly. Below is one of my rough attempts at doing some rocks using clay tubes and trim dynamic.

I mentioned earlier that I have been busy with another Art test for a different company and I thought I would share with you what I did and how it went. For this Art test I was set a challenge to replicate Ground based off a image. This task was very similar to what I did for TT Fusion but had a different challenge based on perception of coverage, species type and height variation. I choose to use a wild grass section from the London Meadows which you can see below as I had the freedom to select my own reference picture this time around.


I first started my challenge by really studying the structure of what plants made up the characteristics of the environment, whether this be by shape, colour, clustering or height. I had to ask myself what makes this field distinguishable. I wanted to make sure also that there was enough coverage and variety in each height layer in the grass. 

After choosing the flowers and plant shaped that I though stood out or highlighted characteristics of the meadow grass, then began the difficult task of trying to find out what species of plant they were to get correct references and pictures to work from. Spending time researching stuff like this is pretty vital for trying to replicate something. you can never have too much reference to work from as it helps breath believability into your work.

I then went about the task of modelling the individual plant pieces in highpoly which I could then use to kit bash plants together once I had baked them down with render to texture.


Once I had modelled the bits of the plants I wanted to recreate I then ordered them in front of a square plane. This means that when I do a render to texture I had my diffuse, normal map and alpha map in exactly the same place in a game resolution texture of my choosing (1k, 2k or 4k).

With all my plant pieces arranged ready to bake, I did a render to texture with a super sample pass and 0 padding. This was so my Alpha mask was really clean and would give me the best results when I used it UE4 as a opacity mask. 

With my diffuse normal and alpha textures now baked I created a new square plane in 3ds Max and applied them to it. From here I went about crudely separating each part out with the cut tool giving enough geometry to bend and deform each piece. This would give me a good amount of kit bashing objects to construct my plants with.  


You can probably see from the image how I have used all the pieces in each plant and also reused pieces the same pieces in different ways to create a variety of shrubs, plants and flowers of different heights and shapes. I paid close attention here to my reference image as I wanted to make sure that each piece would help with each vegetation layer of the image. This would help me to build up densities of different plants for each height layer of the the reference photo and not create unrealistic plant heights that would break the believability of the piece. I would look at each piece in a orthographic view next to each other here to aid this process.

In Photoshop I then created a spec sheet for the plants. Using the select colour tools and the all important refine selection I was able to pick out variations of height and details from my diffuse texture and integrate them into my specular map.

Once happy with my spec texture I removed any light information from my diffuse texture. This is so the normal and spec could work well with the lighting of the environment. I find its easier to pick out details for the spec with lighting information still in the diffuse texture, but you can do this at a earlier stage if it suits your style of work.

I did a quick detail pass using NDO to enhance my textures I had baked then brought my models straight into UE4. I created a master foliage material, terrain blend material and static mesh material so I could create instances for each object so I quickly change material properties without having to rebuild the whole material to see what it was doing in the engine. Doing material adjustments in this way is just an all round time saver and you have mode control of the over all result.

After creating a new foliage layer, placing each plant piece and painting a terrain with some blends I got pretty much the final result. 




I broke up some of the plainer parts with quick rocks I created with ZBrush and Substance Painter to add a little more character to the piece. This was also show off my organic baking onto solid objects even if they were simple ones. 

I was pretty pleased with the over all outcome but reflecting on the final outcome of this I could have done better with the grass by creating them from photo reference rather than modelling each blade of grass in Max. This would have helped the piece by keeping everything at a photo real level rather than having a stylised bits and some realistic. I could have also matched the green saturations better with the reference photo for my submission. This would have been a easy change with some simple node manipulation in UE4 but I wanted to show what I submitted rather than me changing it to be perfect. 


This pretty much covers my process for creating this piece and I hope this helps anyone who is looking to create foliage in the future using the same pipeline as me. All in all, it took 2 and a bit days to create. I am happy to say after a few interviews and this art test that I was successful in my application and am now employed again at Playground Games in Leamington Spa. This has seriously made my year getting this job and I'm literally crapping myself with excitement to work with such a prestigious team of artists.

Wednesday 16 December 2015

My art update for this week

Hi all,

I know I have been a little rubbish with my consistency in posting up my work due to art tests. I though it was about time I shared with you what I have done for one of them at least. Unfortunately I didn't have my application furthered with this piece and I will do through what I thought went wrong with you as well.




Above are some of the renders from my latest art test. If you have been following my blog for any length of time you will know that I am mainly a hard surface modeller. When I received my task and noticed that it was organic I did crap myself just a little. As you can see certain things turned out pretty well whilst others looked pretty weak.

I guess I will start with what turned out good... The loose rocks and foliage I feel turned out really well. After refining a pipeline for these they were quick and easy to reproduce. I used this pipeline for a later art test that I am still in the application process for so I will share that after I have had a decision made whether I have the job or not. The broken pillar attached to the wall turned out ok (not brilliant but passable) as well. I am pretty rusty in Z Brush and really I have no excuse for not using it more. I think it could have definitely turned out better with more detail in the high poly model but my skills aren't good enough yet ( I am spending more time in Z Brush now to up my skill level) to produce the detail required for a better model.

The wall and the tree is where My piece was really let down. I realised doing this that the tree should have been UV'd better than the method I choose to go down which would have enabled me to do normals to define the shape of the trunk more. It ended up being pretty flat and should have had more character. The lighting of the tree leaves and branches didn't quite turn out right so I will definitely be looking into that further too. I tried to use the slide normal thief plugin to correct the lighting of the foliage as a organic body but I needed some more time to learn how to get it right for next time.

The wall... well, the wall was a little bit of a mess. To save Time I tried to do this completely in 2D and NDO and really I lacked the fidelity to get the most out of it. I think creating a Tiling wall in ZBrush would have been the better way to go now by using the same method as how I created my standalone rocks.

I guess from this piece I have learnt a lot about how to recreate Organic objects (I still have a long way to go but as I did this in 6 Days I can say its not a bad effort for a starting point) and can no longer be complacent about not learning Z Brush. I trusted myself enough to craft out some rocks and the pillar and they turned out at least OK. I think a little more time and effort into learning the ins and outs of ZBrush will really help me as an artist from here so I may take a bit of time away from my tiger to learn ZBrush and get it cemented into my work flow.

I think to push my knowledge further with organic objects I will continue to work on this piece until I am happy enough to put it on my Portfolio. This way I won't let the things that let me down stagnate and left forgotten until I need to do them again next time.

That is all for this weeks early post. I will keep you updated as to how I am progressing with my own work as well as job applications as I do them.

Sunday 6 December 2015

Week Update...if not a little late, MWO Marauder.

Hi Everyone,

I have to say sorry again as I have had another busy week making another art test for another games company. It's been a crazy couple of weeks doing work trying to get things done to a standard that I am happy with. I still can't share what I have been doing yet as I am still waiting for feedback on how I have done for my first one but like I said I promise to get that up asap.

In other non related news to work, the new Marauder came out for Mechwarrior online and is the first of the Classics to be released. This came out on the 1st of December (I am a little late with getting this out I know) and I have been having lots of fun with it since finishing my Art Tests and using breaks in-between work to try it out.


Each variant has a good mix to shake it up with a different build set up of energy , ballistics and in some cases missiles too. The mech has turned out to be quite tanky and with the very narrow arm arraignment has a small spread of fire. The stock builds aren't bad either which surprised me as normally you have to strip most of it out and start again (even if this is what ended up happening). I've had some real fun adding triple AC5's on to this and putting down suppression fire whilst energy splurging what I have in my arms to add to the damage.


I thought to keep it short and sweet this week I would share a few of my builds with you so you can see how I have set up my Marauder variants.

Here is my MAD-5D Build or what I like to call the Viking Marauder Medium to close range mech for putting the hurt down.

Now my MAD-BH2 build which is just an energy vomit build but a good money maker.

My MAD-5M which was a experiment with UAC5's

and Finally My Triple AC5 MAD-3R/MAD-3R(S) build.



Besides this mechs tendency to be a bit slow when compared to other heavy mechs its high ballistic mounts and the amount of ballistics this mech can carry make it great for rattling enemy's with suppression and a constant stream of damage.



That is all for now but I will keep you updated on how I am progressing with my portfolio and applications for jobs later this week.


Tuesday 1 December 2015

Week Update

Hi Everyone,

I know its been a week since I have posted up hear last and I have a valid excuse for that as I will explain. I had an art test through for quite a exciting company. I don't want to say which company that is just yet until I have permission but I promise that as soon as I have feedback I will have it up here as quick as possible so you can see what I have been doing.

Sorry for the unexpected silence but I will keep you updated as to what happens there.

Hope you all had a good week and I will see you back here late week for a catch up of what I have been doing in the job search/portfolio.