Murk

In the last email from Dave, he suggested adding some kind of murkiness to the water. Joey was able to find a quick and effective method by using a volume cloud and applying displacement to that cloud and having that underneath the surface of the water to get the effect of mud or or some kind of algae growing inside of this bath. I think this helps the water a lot and it brings it to another level of interest to this asset, especially since this is taking up almost the whole frame for a good portion of our camera move. 

The water is at a good place and this is really close, if not definite, to what we will use in our scene.

Gone Gobo

We took out the reflection of the gobos from the surface of the water, so now it is only reflection the sky above us. There are points in the camera move where is looks flat, and that is being caused by us looking directly down onto the water preventing us from seeing any displacement we applied to the surface. As the camera moves it becomes much more apparent and doesn't look quite so odd. There is also a reflection of the top of a tree that is in our HDR and it is catching nicely in the reflection, breaking up the solid blue reflection of the sky. This tree reflection also moves around the bath as the camera moves through the scene, which also helps give it a little more interest.

One problem that we came across when taking out the gobos in the reflection is that is we turn of "visible in glossy" for the gobos, it effects the entire scene, not just the water. It made all the shaders in the scene seem extremely shiny, and this was not good. So we decided to render the water separately from the entire scene, and Joey was able to comp it together in Nuke to make it work. This was a little bit of a speed bump and resulting in us having to render twice the amount of frames, but we will make it work.

After emailing Dave and Megan about this update, Dave said this is working much better than before. He also said that since the gobos are gone the water appears a lot clearer than before, so he suggested adding some kind of murkiness or something similar to growing algae to help break it up and sell the fact that this is stagnant water that has been sitting for a while.

New Water

I implemented the comments about the water shader from our last session. I lowered the brightness of the reflection of the gobos above us to make it less distracting and confusing. I also backed off the refraction roughness to make it not quite so blurry at the bottom of the pool. Last time, the bottom seemed more out of focus rather than just deeper underwater, so I adjusted accordingly. 

Their feedback on this pass at the water was we should take out the reflection of the gobos all together. It is just getting too distracting, even with the lowered brightness, and in reality there wouldn't be that many branches directly above the middle of this space anyway. We should just let the sky above be reflecting into the water.

Water

After the feedback form Dave and Megan about the water, I decided to take a stab at the water shader since Joey was working on it last week but he seemed to be doing a lot all at the same time, so I offered to try the water. I took out the subsurface scattering and added roughness to the refraction to get the effect of the water blurring the bottom of the bath as the water got deeper. I also adjusted the color a bit because it seemed to blue and too clean for standing water in this type of setting. 

The comments that were given after this adjustment were that this method is working a lot better than before. The reflections of the branches in the gobos above the water are really bright and are quite distracting, so I should try lowering the brightness in the specular. 

New Steps!

These are my most up to date steps and circular wall, being shown in the scene instead of the light rig. This time around I added a lot of large scale color variation to the stone texture. I tried being more logical with where the light and dark spots are, having indentations being darker and protruding spots lighter. For the growing moss/lichen, I added a lot more of this as well, covering a good portion of the steps as a whole. There were two different kinds of lichen I was going for; the the more solid, pale areas and the bright green, more splotchy areas representing thicker moss. The pale green areas are lichen that grows quite thin and close to the rock, and I wanted that to be layered underneath the heavier, thicker moss that grows on top. I also tried being more logical with how this is placed as well, having more concentrated areas growing along the bottom and along the seams and corners where blocks would meet, rather than just on exposed faces. Some patches connect with each other, or come close, creating larger areas of lichen instead of having a bunch of small, isolated areas. 

These same methods are applied to the circular wall behind the steps, although I think the stone texture is a little too low-res, so this should be addressed in the coming days. After painting the textures, I brought them into Photoshop and adjusted the brightness/contrast and the levels to help bump the contrast because they were coming across very flat in the scene. 

Dave and Megan seemed to like the new textures for these assets. They said the lichen and moss are working well and it really gives the sense that it is growing on top rather than just painted on. So they next step is to paint the other set of stairs so we have at least two different sets, because right now I only have one and they are duplicated around the scene. So I will also be addressing this in the coming days.

TX Manager

Our renders are getting extremely long, getting the 2+ hour kill on the renderfarm, and we need to figure out a way to optimize our scene and our textures to get this time down. Our biggest point of optimization is using tx files for our texture maps. This allows the renderer to use only the biggest map size it needs relative to the camera, instead of using high-res maps throughout the whole scene when most of them aren't necessary, especially towards the end of our camera move since everything is fairly far away from the camera.

Our first attempt at converting all of our textures to tx files, it did not really work correctly. It was not finding all of Joey's UDIMs for the lotus bath, and as it was going through all the textures it always got stuck on a certain texture and wouldn't move past it. So after many tries and asking other groups about it, we found out that the tx manager in Maya 2017 doesn't work. So we tried it in Maya 2016.5 and it worked perfectly. The only other thing we had to keep in mind was taking out the unpremult flag in the tx manager everytime we wanted to do this because it would gamma correct everything.

Scene is Coming Together!

This week was the first time that all of our assets with their textures have all come together in the whole scene! Some overall changes were Katie has been updating here tree textures, Joey has been adding and adjusting the scattered leaves as well as texturing the floor and his lotus bath, Michael has been texturing the roofs and arches, and I have been working on my steps and walls. Joey added some water to the lotus bath as well. Everything is working well together so far, but now that we have everything in the scene we now realize that the lighting is much too dark now and it is hard to see our details and textures. 

Megan gave us feedback on our textures, and her biggest point was that there needs to be more variation in the stones because right now they are feeling just too uniform and all one color. They also noted that it is hard to judge the textures in this lighting since it is so dark, so once we adjust the lighting they can give us more comments on them. For the water, they were saying this it is way too murky right now and there isn't any reflection so its looking really flat right now. We did want the water to be murky, but this water we have isn't that deep so it wouldn't get as fuzzy as it is here. Right now the murkiness is being driven through the subsurface scattering, and they said that they would not use SSS and try adding roughness to the refraction instead and see how that works. 

Revisited Steps and Walls

This week I was able to go back into Zbrush and roughen up the corners and edges a bit more on the block for the steps. I also applied this method to the wall behind but it is not shown here in this image. I also adjusted the texture of the stone and made it more of a green-grey rather than a blue-grey to better match the rest of the stone that Michael and Joey put in the scene. The light spots in the blocks are more pronounced than they were before. For the lichen, there is more varying color to help make it not look so flat or as if it is  painted on rather than growing. At this stage, the use of the lichen is in the process of being applied to the circular wall behind the steps but has not yet been executed in this image.

The feedback that Dave and Megan gave me on these assets are that the color of the stone is working better, but there needs to be more large-scale varying color in the stone itself still. I have the light spots, but I need to add some darks and and make it a little more random so its not so uniform and one color. The lichen is also working well so far but there needs to be a lot more it of. Dave also suggested adding some vertical dripping patterns in the lichen as well, is appears a lot in the references we find. 

Still a work in progress!

Re-textured Steps

For this next iteration of my steps, I was not able to go back and change the sculpting of the shapes in Zbrush. But I was able to start the sculpting of the circular wall behind the steps. They are not quite as chipped and rough as I would like, but I got a good start on them, as well as getting the basic texture.

In regards to the texture of the steps I tried adding some light spot in the rocks from where areas chipped off to add more varying color to them. I also added a layer of moss/lichen that has grown on these old steps. This is just an experiment to see how it would work, I plan on adding a lot more than what is shown here. Once I produced the layer of the lichen, I used that same map and converted to black and white and used it as a bump map, to add some height and thickness to the lichen to help prevent it from looking as if it was just painted on.

The comments from Dave and Megan were that there still needed to be more large-scale varying color in the stone. The light spots aren't coming across enough and there should be some dark spots as well as light spots. For the lichen, it is working well and I essentially have the go-ahead to keep working with it. There should be some varying color in the lichen as well, right now it is just all one color. These techniques for the stone and lichen still need to be applied to the wall as well.

Updated Steps

After the comments I received from Dave and Megan last week, I revisited my steps by changing their shape and also added a texture. For the shape of the steps, I kept the edges and corners sharp (they were too rounded before) and attempted more realistic irregularities in the faces (last week they seemed too lumpy and rounded). I also changed the shapes of the individual blocks, making them more varied in size and shape rather than being too uniform. I tried adding more rough edges and chipped sides, but I don't think it is coming across well enough and should be pushed further. 

As for the texture, I used an image of dark sandstone which is the kind of stone used to construct Angkor Wat that we are taking reference and inspiration from. I want the stone to feel similar to what is used to for the lotus bath and the roofs of the covered walkway, but not too close that they blend in. 

Some of the comments from Dave and Megan are that the stone could be more severely chipped and roughed up and that what I have now is not prominent enough. In regards to the stone texture, Megan commented that there needed to be more variance in color on a larger scale. Looking at the reference photos, not all the stones are exactly the same color, so I should add some light and dark spots.

Updated Shader Balls

These are my updated shaders based on the comments that Megan and Dave made during last week's session. I didn't get to all of the shaders, but I adjusted the ones they made the biggest comments on. These shaders are (in order); gold, hollow glass, olive oil, rubber, and the extra rock shader that will be needed for our scene. For the gold, I desaturated the color and adjusted the specularity. The glass was made hollow, it was just solid glass last week, so I extruded the edges a little and adjusted the roughness of the refraction and specularity accordingly. The olive oil was also desaturated in color as well as adding more roughness in both refraction and specularity. The rubber was adjusted by making sure the specular weight is at 1 and the fresnel is driving how bright it is, as well as increasing the roughness and adding bump map to help break up the surface a bit. My rock shader was adjusted by making my own texture map and layering the white splotchy/splattering spots to try and get the look of old dirty stone.

Overall I think my shaders are much improved since last week, and Dave particularly liked my hollow glass shader.   

First Attempt at ZBrush

This was my first time officially using ZBrush to apply displacement to the steps I am making for our scene. These stills are from the turntable I made for our session with DreamWorks. During this process I learned a lot of how NOT to do displacement in ZBrush. Some of the things I  did wrong when starting out was not prepping the geometry correctly in Maya before bringing into ZBrush. I did not realize that not having some kind of beveling and more than one face on each side of the block was problematic when applying the displacement map to the geometry back in Maya. Then I did not realize that even though I beveled the blocks, when I add more subdivisions in ZBrush, the polygons within the beveled edges get tiny and become a problem when smoothing and laying out UVs afterwards. So I learned the hard way, that you should bring in your initial geometry from Maya into ZBrush, use a deformation tool to smooth, then export out and bring back into Maya, layout the UVs again so there is no stretching, then bring back to ZBrush to apply displacement.

Once I better understood the steps needed to take before starting to sculpt, and also gaining a more comfortable method of actually applying displacement through several kinds of maps, I became more confident and comfortable in making these stairs.

After a taking a break from looking at these stairs for so long, I came to the realization that the individual blocks are too rounded on their corners, and should be much more sharp and jagged. So going into the DreamWorks session I already knew that might be one comment they would give me. But another thing they said was the blocks themselves should be more varied in size, a little more random in the way they are assembled, and not so uniform.  

One thing that I came across while researching stairs or steps in ancient ruins was that they are, for the most part, fairly well preserved. So when making these stairs, I'm not sure how far to go with the destruction or how rough I should make them. Dave commented that even though realistically they might be pretty well preserved, I can take some liberties in adding more cracks and roughness to help sell the "ancient ruins" look of this place. 

I have a lot to improve on after hearing these comments, and I look forward to advancing on the look of these steps.

First Pass at Shader Ball Exercise

These are my shaders that made it to the quilt that represented the shaders for our team. The shaders are (in order of appearance); glass, gold, matte plastic, olive oil, rubber, painted metal, and an extra shader of rock to be used in our actual scene. The comments that were made on these shaders we the glass should be hollow (something they forgot to mention to us, the gold should be a little more desaturated, the matte plastic should have more roughness, the olive oil should have more roughness in the specularity and the refraction as well as being more desaturated, and the purple rubber should have more roughness. 

The comment that Megan and Dave to the class as a whole was that the specularity weight for all materials, whether metals or dialectric, should be at one and the fresnel level should be the value determining how shiny it is. This was unclear to me before this session, so after these comments I will revisit all my shaders and make sure they all follow this rule. 

Light Rig Tests - Something is Wrong

Our first attempt at making a vdev light rig test did not turn out the way it should have. In dealing with the color management of our renders, our first thought was to keep everything in raw to have a linear workflow. We imported the HDR as raw, set the color management settings in the preferences to apply output transform to renderer, and set the viewport settings to raw. But after taking all these precautions, it still didn't look right to me, the whites are too blown out, the blacks are too dark, and the black plastic sphere is not showing any reflections. We tried everything we could think of but I still wasn't convinces that this is the way it should look.

So after some help from the class and some feedback from Dave, we found out that the viewport needs to stay in sRGB gamma, not raw, so see the correct results. So now we are on the right track to making a proper vdev light rig.

04/14/17 - DWA Session

Updates: We implemented the mandala concept into the design and layout of our scene. Each component of our space is alternating shapes of circles and squares. We chose a lotus shaped ritual bathing pool to be at the center. The symbol of the lotus is also used in the design of the four arched entrances into this space, and will continue to be used throughout the architecture as we move further with development. Beyond this immediate space is another perimeter of covered walkways, keeping with the radiating pattern of the mandala. A few banyan trees will also be implemented, with their roots wrapping around our architecture, with a few reaching towards the center lotus pool.

Feedback: The space is really interesting. Our current camera move stays within the space, but Dave thinks we should show off the space that we designed. Our initial camera move begins with looking straight down at the lotus pool while rotating, emphasizing the symmetry. Dave suggested continuing with that motion all the way up, until you can see entire space. Or slowly moving to one side as you move in height to end up with a 3/4 view of the space (like the picture above). 

As for the camera move we currently have, the transitions need to be smoother. It feels as if it is 3 separate moves. It is a really long shot travelling a long distance, it can be done but needs to be finessed more. 

The Mandala Concept

Thankfully I am taking an architectural history class right now and we are learning all about ancient civilizations, more specifically ancient Buddhist and Hindu cities. Something that is found in almost all the architecture of these cities is the use of the mandala concept. They layout their temples and cities in the design of this mandala, essentially consisting of alternating levels of circles and squares. The squares usually represent gates or passageways to the next level, with these four entrances oriented to the four cardinal points. 

So when we were thinking of how to design our new space for our project, I thought why don't we use this same mandala concept. We are already taking inspiration from cities that use it, finding more reference will be great since it is used so often, it offers a higher significance to the whole space, and its also results in an aesthetically pleasing design.

We also decided to change the well at the center of our space to a lotus-shaped ritual bathing pool. Lotuses are also very common symbols in these ancient cultures. Also ritual bathing pools are much more significant than a well in these cities, so it will be more fitting to have this at the center. 

Back to the Drawing Board

After our session with DWA on April 3rd, we needed to go back to the drawing board to redesign our scene. We knew we wanted the foreground to be our focus, and we first thought of having a well at the center of our scene. So we reconfigured our whole setup and had a new layout for our environment, but it still didn't feel right. We needed something more interesting or important than just a well in the middle. So we asked ourselves the question of why aren't these people here anymore. We felt we needed to show that something happened here and this civilization could no longer live here. 

So what we came up with was an earthquake. There was be a large crack in the ground, essentially split the space in two. The shifted ground would go through the well, having half of it fall into the crack. We thought this could give us a way to move the camera to reveal the space better and have a little bit of a "wow" moment with the crack.

But after getting several people's opinions on the earthquake, we decided to abandon it because in reality, if there was that big of an earthquake, then the whole space would probably be totally collapsed and there would be almost nothing recognizable left. 

We came to the realization that we needed to completely redesign our space from scratch. We have to design like we were the architects and it has to be built for a purpose. We can't just try to make things fit together that don't really belong. So we need to go back to the drawing board, again.

04/03/17 - Pitch Modifications

Updates: We got rid of the archaeological site, just focused on the ruins themselves. Decided to make a well the focal point of our scene. We want to show that there is more to the city that we aren't seeing, so we put a large important building in the background (the Observatory). Still would like mountain range in the far background and local fog settling in around the scene to keep that mysterious feel. We also introduced come banyan trees that have long roots that wrap around whatever is near it. We thought the tree could help push the point of this city being really old - having a full grown tree growing on something it shouldn't be.

Feedback: The foreground and background are contradicting each other on which is the focus. We want the foreground to be the focus but the background in our concept art is pulling focus from foreground. So we should just pick which one we would like to emphasize and run with it. It doesn't really matter which one we go with, but just pick one and stick with it.  

03/31/17 - First Pitch to DWA

Synopsis and Concept: Two archaeologists are setting up camp early morning outside of ancient ruin city in the South American mountains. One of them leaves to get more supplies and when they return, they see things knocked over and the camp looks as if some kind of altercation has happened. The other archaeologist is gone, but there is a trail of drag marks leading away from the camp and towards the city, as if he was attacked and forcibly dragged away.

The mood is mysterious, using early morning light and atmosphere/fog throughout the scene. There is a contrast of modern day equipment from the archaeologists and the ancient city in ruins.   

Feedback: Get rid of archaeological site, just focus on the ruins. The setting of the ruins is interesting enough to tell a story, don't need to add an event to it. The mysterious mood to good, stick with that. The story of an ancient civilization is mysterious already; what happened to these people, why don't they live here anymore, etc... Let the environment say everything for us.