9 l 6

Quite a lot of progress has been made since the last update on creature progression so be prepared for a long post xD

After building and skinning the muscle rig it was time for the fun stuff. (more sculpting and texturing) Initially I had a few different ways I could go about doing this. I know I wanted to go further in the sculpted detail but with little time left I didn’t want to painstakingly sculpt in everything from scratch. Instead I looked once more to Javier Blanco’s pipeline and decided to take his approach. Which was building a sculpt driven by displacement maps stitched together in mari. For this option I needed displacement textures of various animal skin/scale textures, and lots of them. So I had taken my lecturer Steve’s advice and used CrazyBump to turn my texture library into useable displacements.

Once I had projected on my displacement textures where I wanted that detail on my creature, I then exported all the maps. From here I applied the maps as displacements in zbrush. Bam, instantly I had something I could work with, it was no where near perfect but it gave me a base on which I could build on. 

From here I spent a day polishing the sculpt, I could have spent a lot more time detailing but I really had to move on to the texturing.

I gathered some references of the colours and patterns I wanted for this creature, such as these below-

From here I went back to Mari, this was the first time I had used Mari to texture, so I didn’t really know what to expect. I watched a tutorial to help me set up my scene and then had various tutes playing in the background while I worked away.

These were some of the ones I found useful-

http://www.digitaltutors.com/tutorial/1481-Combining-Painted-and-Image-Based-Textures-in-MARI

http://www.digitaltutors.com/tutorial/1603-Essential-Strategies-for-Hand-Painting-Textures-in-MARI

http://www.digitaltutors.com/tutorial/633-Combining-Painted-and-Image-Based-Textures-in-MARI

Firstly I painted a base texture, which consisted of various greens. One of the features I loved about Mari was the ability to drive masks made from channels. I used the maps that I had exported from zbrush (normal and displacement) to drive these masks. The control that you have on a channel mask is also amazing, you can play around with the curves and intensity to see in real time (if you haven’t baked your paint in) and see exactly how that affects the texture you’ve just painted/projected on. Mari also has the ability to calculate the AO of the model by the click of a button. You can then use this calculated AO as it’s own mask as well, with all the control of it as well. There were also many other procedural mask types that I played around with such as noise to get some variation in the pattern, colour that I was painting.

I used this techniques to hand paint most of my model, I also used a mixture of textures in places that I wanted to project certain details, such as the eyes, and also the bony areas of his body which I then tidied up by painting over. There are also some nice warping tools that you can use to really put the projected texture in places that you want it. For example I used this particular texture for his bony plates on arms and legs, but of course it wasn’t going to match up exactly however using the spline warper tools I could bend it and pull it to the shape I needed before baking it in.

One more thing that I used quite a lot throughout the process were the layer adjustments and adjustment stacks. For instance, I added a multiply adjustment to a layer which I would then drive by a mask from my displacement map or AO to get some nice dirt or darker colour in the crevices. I used this on the claws and spikes and then painted on the mask in places I wanted more dirt (like under the claws or where they meet the skin). I also ended up with the reddish colour on his bony head and back crest by accident, I was playing around seeing what an overlay of texture would look like with different blend modes like screen/lighten/colourburn etc. However when I hit contrast it made the places where I overlayed the texture and earthy reddy colour with tinges of purple in some areas. I really liked the look this achieved and the best thing about it was that it was completely unplanned.

After I had finished the diffuse texture, I also painted some other maps, like specular, gloss, reflection and emissive. While watching this tutorial, he mentions that we can fake subsurface scattering with an emissive map, I hadn’t thought of using emissive maps this way before and was interested to see the results it would yield. So I painted with a pinkish colour on a very low opacity in places where I wanted this, like on the cheeks, belly, fins, palms and mouth area. Once I had made all my maps I exported them out so I could compile it all in max.

My material setup was pretty complex as I had made the decision to use multiple texture tiles. I used composite maps to link up all my uv tiles and adjusted the U and V values on the bitmap parameters for each layer of the composite map.

 I did however encounter an issue with my displacement maps, when I exported them out from zbrush it exported all my subtools as different exr files and wouldn’t let me merge them in zbrush. So I had to use nuke to combine those that were sharing the same uv tile. I used merge nodes and then re-exported them out.

Once this was done, I went back into max and continued to setup my shader. Once this was done, it was time to play around with the displacement parameters. I wasn’t sure how to set this all up but I found this useful tutorial on that. It seemed like a really weird way of setting up the materials but hey, it worked so I’m not complaining xD

After doing some test renders I came to realised that my shader didn’t really seem to be doing anything, I even cranked up the specular and it made no difference. Then it occurred to me that because I was using vray to render, I also needed a vray material not a standard one. *facepalm* So I quickly changed that and reconnected all my maps. However now I realised I faced another issue. The vray material uses different parameters and instead of specular I needed a roughness map, along with two different types of gloss and reflectivity. At this point I went back into mari to paint some more maps.

 I made a roughness map, and also adjusted my reflection map to suit vray, at this point, I decided to also make a bump map (using procedurals and painting) just to bring out those little details and make my reflective and gloss maps really catch that. I wasn’t quite sure how to setup the vray shader to get the wet glossy look I wanted for the skin of my creature, as there were multiple different slots for reflection and gloss. I did some googling and found this setup that they made to make a wet road. So I plugged in my maps according to what they had done and it worked!

I did some test renders and continued to make small adjustments to my maps in mari until I was happy with the results.

One of the main sources of inspiration for both our final project (CIU212) and studio project (AMN230) was Guillermo del Toro's film Pacific Rim (2013). This particular film is not only iconic but the complexity and level of detail that ILM managed to produce made it an exemplar in the VFX and animation content that we as a group are aiming to produce.

The film is set in the near future, where a portal has formed at the bottom of the Pacific Ocean, in which waves of Kaiju (Huge monsters) emerge to wreak havoc upon the earth. In order to survive, humans have developed Jaegers (Giant Mecha) who are piloted humans in order to combat the Kaiju.  (Gilchrist, T. 2013)

The prospect of ILM creating giant monsters versus giant robots for Pacific Rim was something many film goers were hungry to see. Perhaps aware of that anticipation, and knowing the challenges that lay ahead in realizing the enormous 'Kaiju' versus 'Jaegers' fights on screen, the studio retooled many elements of its effects pipeline for the show, including simulation, lighting and rendering. (Failes, I. 2013)

As part of my research, I watched many “making of” videos on pacific rim (among other titles) over the course of the trimester gathering what insight I could into the processed that went behind making this incredible film.

Some of the Pacific Rim ones that were particularly relevant to my role were these two  -

There were focused on the creation of the Kaiju. The design processes that were involved as well as how they tackled scale and creating believability. As my main role in both projects is to create the Kaiju sized creatures, these were some relevant pointers that I gained from watching the videos-

Looking to existing animals for inspiration, based on life forms that existed on earth.

Creating their personalities, taking time to develop how the creature ticks, and knowing how they would react in certain situations would further the continuity and believability of the creatures.

Working with scale, -the textures have to be very detailed, even the individual scales are made up of bumps and cracks.

I tried constantly throughout my work this trimester to incorporate this into my pipelines. I gathered plenty of references all throughout the process, whether it be for the initial concepts or for further detailing, such as skin/scale detail and even using these to make maps to drive my sculpt or textures. I wanted my creatures to be believable so I went and created a bio along with making a scientific name for one of them. I have yet to make one for the other, but I will in the near future. I think working with scale is something that still needs to be addressed on my behalf. Especially when it comes to animating. Because of the size of the creatures, the speed and forces of movement changes. I have found some sources of reference that I will be referring to when it comes to practice my animation for final project but I will definitely need to do more research and testing in this area shortly.

References-

Failes, I (2013). Monster mayhem:Pacific Rim. Retrieved 7th May 2015 from http://www.fxguide.com/featured/monster-mayhem-pacific-rim/

BEHIND THE MAGIC COLLECTION: THE VFX OF “PACIFIC RIM (2014) Retrieved February 27th 2015 from http://www.andrewblodgett.com/watch-behind-the-magic-collection-the-vfx-of-pacific-rim/

Throughout our CIU212 final project we outlined the kind of project management practices we’d be using. For communication we’ve mainly been using google hangouts however we’ve come to realise that is is nice tool for chatting, but not so great for managing a project as things can easily get lost and then you have to scroll through an endless feed of spam to find it again. So we’ve decided to switch over to slack from now on for managing the project. I’ve used slack before and found it to be a really powerful tool in this department and I think it will be suitable for our project. Another aspect we are going to have to address is file management. At the moment we are using google drive, which is all well and good, until we reach our data limit, which unfortunately will be relatively soon. At the moment we still have to discuss our options to this dilemma and hopefully come up with a solution in the near future. Our filing system on drive has been straightforward and I haven’t encountered any problems so far. We’ve kept it reasonably clean and made relevant folders for all our data whether it be documentation, presentations or production assets. The team has been a pleasure to work with, and we seem to be on the same mindset when it comes to making project decisions which is nice. So far so good with the team, I’m not killing anyone yet.

As for the project in general, we’ve decided over the last few days to implement a major shift in our delivery. Originally we were going to render out all our animations and simulations from 3DSmax using vray, however it occurred to us that rendering times may blow out and considering we are quite limited in our timeframe we’ve decided to branch out and consider real time rendering engines such as Unreal 4. I have had previous experience in using EU4 and really like it. It’s very powerful and I believe we would be able to get the quality we want out of it and being able to make changes in real time will obviously save us a lot of wasted time that would otherwise be used on test renders. There are a few hurdles that we’re going to have to cross if we choose this path, some of which include how we would do certain things like ivy generation, however we might be able to bake the animation in and export it that way. Over the break we’ve all decided to do our research in this department to see whether this would be a viable option.

On a side note, here is a drool worthy creature made by Eric Wilkinson available for a demo download, in UE4.

As some of you might know, I have recently begun live streaming some of my 3D work on a site called myminifactory.com (myminifactory.com/tv/list) I believe it’s now out of it’s beta stage and, there has been many awesome changes happening recently. One of these is that you can tip whomever you are watching if you wish to. I was pleasantly surprised when this actually happened to me last week! This goes straight into your paypal after you add your paypal email address to your account. There is also another neat feature, which is a chat bar next to your stream. Here you can interact with whomever is logged in and watching you. It’s also great to be able to watch someone streaming and be able to interact and ask them questions about what they are doing. This could be a really good way to learn processes and workflows that other people use for creating 3D assets. Overall I feel this is a great site to utilise no matter what kind of 3D work you do, it is however focused mostly for 3D printing however they encourage all sorts of 3D work to be streamed whether it be for animation, games etc.

Another cool feature that I might try and apply for are the design jobs and competitions that they have going on. Here you can fill jobs to make objects for 3D print and get payed for it. There seems to be quite a few jobs going around and many of them seem quite enticing.

Earlier on in this trimester I was a part of a bigger group project as part of studio 3. In this project we were to create an efficient animation according to a flexible brief. My role in this project was animation lead however I did find myself also contributing to some project management tasks along with asset creation.

We used slack as our main tool for communication, personally, I really like slack. I find it intuitive and easy to use. The different channels you can make help to keep it organised and it’s so easy to give feedback and speak with other team members. You can also implement other apps into it like google drive. It also gives you live updates on your phone or desktop if you have the app installed. I’ve adopted this site for other group projects and plan to use it a lot from now on. However I did find that for this particular project it was a little harder to manage using only slack. As I became aware throughout production that only some team members were actually using it. So instead we’d have to not only post things on slack but then also go over things again during class for those that hadn’t been aware of what was happening. This didn’t bother me overly much however it did make communication a bit slower. I feel that file organisation could have been a little stronger as well. We used google drive to store all our assets and pass them on to others in the pipeline however there were quite a few folders with names that were similar making things a little unclear on where people should upload. I noticed this happening when people were losing things and couldn’t locate assets. I probably could have gone in and changed the setup at this point however we were already most of the way through production and I didn’t really want to cause any more confusion.

As animation lead I was responsible in delegating and reviewing the animation that was produced. I did however receive assistance on this front from two of my other team members, Karen and ani301 (Amy). We assessed all the animations and gave feedback and guidance where needed. It was good to have other eyes also assessing the animation as sometimes one person can easily miss things.

Looks great, all flabby and jiggly. Did you find the muscle system to slow the computer down much or is it fairly efficient?