new website, kinda

i forgot about this blog and started a proper site a while back. if your interested go check out my portfolio at http://www.mattluxton.co.uk/

Blimp progression

Blimp progression
along side making the apprentice character i took on the task of making the Blimp exterior and interior. the blimp designs were done by Simon and the Harpoon was designed by Roy and i set about making a low-poly version of these assets, so we could reference in the high res version in at a later time in production. my aim with making the low res blimp was to be able to re-use a lot of the parts for the high resolution version, but there were a few main pieces such as the engines and certain parts of the hull i had to completely remake to a high polycount due to camera close ups. it took me about a week or so to make the low res version of the blimp, i could then start thinking about the interior shots. Simon had drawn up a few concepts of the inside of the cabin and provided me with some nice angled shots to work from. i’d have to say the most annoying things to get right would be the scale. it took us a few attempts to get the scale to work correctly  for everything in relation to the characters height.

90% of the cabin was made by myself, and the rest was a combination between matt and simon. this was something i really enjoyed getting stuck into as i had scope to add little details and lots of pipes and fittings to the room. But as always time is always biting at your heels and you have to make decisions in certain aspects to cut out. The overall aim was to make the room full and feel as if its being used by the characters, thus meaning i had to create lots of assets for this segment of the animation.  this provided me with another great learning spree, from uv space and size allocation to polycount depending on distance to camera. What i really like about working on this room is the pipeline we had set up. i would model the props and unwrap them, roy and simon would then texture the props i gave them and then  roy would organise all the files in which i could plug back into the texture files. this worked really well and work was continuously being produced and completed at a good rate, it felt like we’ve come close to a proper production line.

It soon became apparent to us that we had probably spent far too long working on the interior of the cabin, the majority of textures had been completed and essentially all the integral props were done. we hoped we could come back to some of the unfinished things later on in production if we had time, but my priority next was to focus my time on making the high resolution version of the blimp.
this was an area I was really looking forward to as I thought I could showcase a bit more of my skill and having made the low res version I already knew about any problem areas or things needing to be fixed. My aim was to try and keep the polycount to a reasonable level and I knew I didn’t want to smooth the blimp as this would only add to render time. So what I tried to do was model the blimp to a level that looked as good as the smoothed version but with a lot less polygons, and simon seemed pretty pleased with the results so I’m happy. Any areas that would be close to camera would be my main priority to get looking good. So this would be the propellers/engines and generally the front end of the ship including the harpoon. As a modeller I take it upon myself to add a little bit of flare or personality into the work, so what I did was tweak a few areas of the designs a little bit. Sometimes this was to reinforce themes previously shown and sometimes it would be to make an object look a little bit more interesting. Examples of this can be found on the side of the engines, where I added a few pipes and on the chair of the harpoon gun, I brought the hunter theme back into­ the picture and added some horns to it. A tiny change it maybe but I think it adds a lot to a shot.

Once again I was only modelling and unwrapping the parts of the blimp, Simon took on the task of texturing the blimp and Olly was responsible for the shaders used on it. Overall im really happy with the outcome of it and I think everyone else in the team is aswell. Obviously there are areas for improvement, one being I could probably get away with lowering the polycount in a few areas such as the railings on the upper part of the blimp, essentially just be more efficient in my work flow.

Upon completion of the blimp I became a 3D generalist for the team and essentially a guy that fixes the broken things. I worked closely with the animators and helped them with anything that they needed for particular scenes. A lot of my time was spent on the characters weighting which I mentioned earlier in the report, but there was still work I could model in the mean time. Seeing as the film is predominately in a Jungle we required a lot of plants and trees. Matt and Emma had already provided a lot of foliage, but I noticed that there we certain types of leaves and rocks that could be used to fill space and be used to hide a lot of things. These were relatively easy things to model and texture and I made a variety of each, including different colour combinations for the textures, these were then handed out to Matt and Simon to help fill the scenes. At this point in production I was told that we had to stop modelling things and get on with finishing anything we had made previously and just focus of tidying things up.

So my next assignment was to layout the mushroom scenes which involved 8 camera shots. My aim with this was to create a dense environment, or at least give the impression of a thick jungle for each camera shot. As stated the scene has 8 cameras, so what I did was cut up the set into 8 sections so essentially what you see on camera is what you get. By this I mean that I had 1 set which incorporated the whole jungle, and for each of the mini sets I cut anything away that wasn’t seen in the camera as it only would of added to render times.

It took me a good week or so to layout the scene to a standard I was happy with and then texture all the things that needed doing such as the crashed blimp site and the ground planes.

I had quite a few issues with the scene I was working on, at some point I think the file corrupted and would crash every time I tried to render anything, plus the massive scene file didn’t help with the matter. I managed to fix the issue, but the size of the files per shot were large to say the least. Something I should have changed during the layout was to instance the assets rather than duplicating them directly as this would have saved a lot of memory. Unfortunately it wasn’t until after I finished the scene that I thought about doing this.

I’d now like to talk about the post production part of the pipeline, Tom and myself appointed ourselves as render monkeys and set about trying to get shots rendered as fast and as soon as possible, we also helped with the lighting in a few of the shots. Olly did 95% of the lighting for the whole production, but due to easter breaks he was away from the studio so we had to take over until his return. I’d also heard about ways to render outside of maya using the command line, so I ended up researching into this and teaching myself how to set it up. This would take advantage of all cores available on each machine and render without having maya open which would improve render times further. So from about late Feb-March myself and Tom would start rendering any frames we could, ready to be comped in later.

Due to the vast amount of work to still be done, a point arose in which we had to decide which shots to prioritise to render, one of them being the flyby sequence at the beginning of the film. With this shot I set myself the task of rotoscoping out the propellers when they intersected parts of the blimp. This intersection occurred due to how we had the passes set up in the composition. I Probably rotoscoped a good hundred frames before Pete told me I could have just used an Alpha channel. At best my knowledge of Toxik is very limited and I cant seem to get my head round some of the issues it throws at me, and I decided i’d pass the torch over to Pete as he knows what he’s doing. I tried to the best of my ability to sort the composition out but at the end of the day Pete could do it ten times faster and more efficiently than I ever could.

So all in all it was one hectic production, we managed to get some really nice rendered shots,my favourite being the flyby sequence. The animators did an amazing job with the little time they had to animated due to all the broken rigs and I think as a team we really worked well together. There were a lot of ups and downs but that’s to be expected. Looking back on the project through the knowledge i’ve gained, I would change a lot of things, especially in the way I approach them.

apprentice progression

my role for the project was to create assets for the film and one of the characters which was the apprentice. the apprentice went through a lot of changes throughout the production, this was either due to it not looking quite how the director wanted it, or due to technical limitations found later on in the production pipe line.

the latest pictures on my blog displays the latest view of the characters head at the time in December, at that point the characters face was roughly done along with the hair. the basics were there, but its wasn’t optimal topology wise so I decided to start the face over again with better flow topologicaly. another reason for doing this was the lack of eyelid mesh available for rigging, and as this was a very key area to get right, it was important to pay attention to it. the mojority of time the audience will be looking at the characters face. It was my job to make sure the character could be rigg ready for production.

Simon decided on the level of detail we wanted the characters to have and whether they should be smoothed or not, this obviously increases polycount and render times. this took a while to figure out and the decision was alot of to and fro. so what i decided to do was in the mean time id set a mid ground polycount and add more topology to the character where it needed it, this included re-topologising the hair i sculpted to a more usable state. U
pon completion my character has 12.5k polys, this number varied a little during the rigging/weight painting stage of production as i had to make the decision to remove the front pocket on the characters dungarees due to the results and difficutly it provided during the weight painting process.

weight painting was an area i deffinately wanted to get under my belt before i left university and i’m pretty happy now that i’ve done it. the process took a few days to grasp the basics of and the rest of the time would be spent tweaking the values. one thing i didnt know about weight painting was the fact that there is no way you can make the character work in every pose, atleast at a basic level with a single rig like ours.  i had major issues with the arms/clavical areas of the t-shirt and dungarees, and alot of difficulty with the trousers. certain poses i’d find worked really well, but then i’d move the character and i’d see a problem, so essentially the aim is to make a compromise to get an overall decent weighting on the character. I’d have to say the most annoying issues i had was with the characters trousers, due to the low cut style and the rig restrictions or lack of. the animators had the ability to rotate the characters knee’s and legs at impossible angles, most of the time you can get away with, but with our character, his crotch area refused to co-operate to say the least. essentially it looks like he has a loaf of bread down his trousers which definitively isnt a good thing. something to take note of is that for the biped characters we used an auto rigger with the help of Scott in the second year. this had both advantages and disadvantages, the rigs could be made at a fast pace, but in the end we would find we we’re inundated with problem after problem. for example we had alot of joints in the stomach/chest area that essentially did nothing and with bad naming aswell made it a nightmare to try and weight.

we made the decision to also use a facial rig plugin rather than using blend-shapes, and in a similar fashion this also had major flaws. essentially we had the same issues with the body rig as we did with the head. The most obvious issue i found would be the eye lids of the character intersecting with the eyeball, this is an area i spent hours trying to fix, and for some unknown reason a few of the vertices on the lid refused to move when being weight painted. i tried a variety of solutions to try and fix this, and decided to go for blendshapes as i know i would be in full control of what the movement would be and the animators would have a good selection of movement for the eyes. making the blend shapes took hardly any time and provided some great results, which felt great becuase finally something was going my way for once. unfortunately the rig reminded me that it was in charge, by off-setting the head by x amount with each blend shape. I found out that this was something to do with the history of the head of the character before it was attached to the body rig. at this time it was too late to change anything as production was steaming ahead so i had to leave the blend shapes behind. We were still left with the issue with the eyelids but i had fixed the weighting on the rest of the face, fortunately Georg reminded me that because we couldnt scale down the eyes as a whole, i could scale them down at the vertex level instead and luckily this solved the problem.

Texturing the character

texturing the character was an area i was really looking forward to do, as throughout the production i was responsible for just dishing out the models with  fully unwrapped UV’s and let simon or roy texture then. now i had the chance to really get to know the process and experiment with a few things along the way. Texturing was another area i really wanted to get a firm understanding of aswell before i left uni. I’d been studying alot of other 3d artists and their workflow to pic up hints and tips and something i thought id try out was making an ambient occlusion map. this map provided me shadow detail and cavity/groove information for the character parts. multiplying the ambient occlusion into the diffuse colour and possible adjusting the hue can create some really good result. this is something i now do for most of the models i make. A big part of the texturing process is trying to squeeze as much information into a tiny space, resolution is a massive factor in this situation and what i hoped to do was use a 2k resolution texture for my model, this provided me with enough space for paint information and this was perfect for 90% of the character. there were some sections that i would have like to have more detail, but the resolution couldn’t support it. in hindsight what i should have done was to paint the texture at 4k and then scaled down the texture when i had finished it. this is the process i incorporated when making the rucksack for my character and this yielded much better results and i could add some really find detail to the model.

To finish off the texturing process of my character i started to look at specular maps, which would controll which parts of the texture would show reflections/highlights and could also determine the colour of the highlights if painted the correct colour. i had a few issues with this and ended up sticking with black and white with greys for the texture which would be controlled by a blin material

that pretty covers the character and overall i’m really happy how its turned out, and it provided me with a great learning experience, the turntable of the final character can be found on my website. http://www.mattluxton.co.uk/apprentice-turnaround/

catch up

its been along time since ive blogged, so i’ll cover everything ive done since. ill go into more detail later on for now i’ll provide a list of things ive done.

• character modelling revisions
• blimp interior majority of props
• high res blimp and harpoon turret
• weight painting
• plant and rock modelling
• texturing
• scene layout

big update coming soon!

ive been swamped with work and had little time to blog about it, mainly due to our disseration but it’ll be over soon and i can start putting up alot of the pics ive promised a few post back and all the stuff ive been up to since.

see you in a few days

The Last Trophy – update

So its a new year and hopefully a great one at that. I’ve been really busy over the holidays preparing and writing my dissertation which isn’t my strong point by far, but its going well, and hopefully be to a reasonably good standard.

The Last Trophy team are going to be back in the studio this monday to continue with the project, we’ve all head a good rest and are eager to get back into things. I’ll try and get some pics up on my blog asap and try get some of my personal work together ready to make my show reel.

I hope you had a good holiday and I’ll see you on the flip side.

oh check this out http://www.youtube.com/watch?v=uhyScbAZrFs Making of Tron, looks pretty sweet. I’ve not seen the film yet, but i really want to now

The Last Trophy – update

As usual I’ve been really busy with the group project and sorting out this blog. I’ve had great fun learning simple html all last weekend and ive really got into twitter too. its such a great platform to branch out and find new people and content.

so what I’ve been up to? essentially ive been filling out the hunters cabin for our project. its providing me ample of time to improve my hard surface modelling and keep an eye on poly count . I’ll post a few shots up of the room so far, its a fair way from finished but the basics room has been modelled and several props completed. What im looking forward to the most is adding all these little details to the room, to make it have a bit more character and life to it. As an artist i think its really important to be able to portray and give character to everything, really set the mood and story behind what you are creating. and hopefully by the time I’m done with the room I’ll have provided lots of eye candy and questions for the viewers to gaze over.

for now peace out, see you on the flip side