Week 10 - (Shaders)

The original learning contract was constructed around experiments related to the quality of lighting an environment. It was built to produce work that would look amazing in theory despite the original thesis point was based on economy and average consumer products. I tried in the past to achieve an economic rig by producing a quality rig and stripping down. It might make more sense to do the opposite to see what I get. In order to understand the economy side of environmental lighting I would try and produce a rig that could compete with physical sun and sky.

AIM

The aim for this week was to try and produce a light rig that would be sufficient to rival the mental ray physical sun and sky. The process would have been solely base on starting on getting an economic render than bringing up the quality. I would also look into how shaders could impact this.

METHOD

Firstly I started off with a base render of nothing but physical sun and sky and recorded a time of approximately 2 minutes. The renders were all done on the same QANTM iMac hardware.

Fig. 1, Physical Sun and Sky render

Here is how the following lighting rigs were then constructed then timed:

Fig. 2, Week 8 result

-         -  Week 8 environment with all of its lights and bounced lights came in at approximately 8 minutes

-          - Bounced lights were stripped were replaced with Final gather, this brought the render time to 11 minutes

-         -  Other unnecessary lights were stripped leaving only major lights (ie fil lights). Roughly 6 minutes

-          - 3 point light system with target lights as fill lights. Still roughly at 5 minutes.

-          Having the directional light on its own with final gather approximated the same time as physical sun and sky.

From here I also looked into shaders and the impact they could give. I went through a handful of shaders, including ciesky, portal light, photometric light etc. It was a bit hard to understand what I was doing as I didn’t fully know the potential of these particular shaders. After a while I was hit with an answer.

RESULTS

The reason all of the light rigs were taking so long is because I was using multiple Mayan lights which were apparently more expensive. For every light that is created, light information is multiplied and therefore having less is better.

Physical sun and sky uses only one directional light and therefore already trumps Mayan light methods.

My experiment with shaders seemed a much more sensible route to take. With my knowledge from previous weeks I was able to get some results with shaders but still got the same render time.

LEARNING TO FAIL

It hit me that if I was building a light rig that had a dependency on shaders that I wouldn’t have much customisation over traits light shadow colour. Therefore this method was no better than Physical sun and sky. Mental ray had a system that had been developed and engineered for years. The sky node and the sun node work too well together as well as other shaders in the mental ray network. I wouldn’t that my efforts had been foolish but I grew an understanding on physical sun and sky worked.

Overall I probably have not really travelled far but have gone through a long journey to understand the value or Maya/Mental ray light systems.

   

Week 9 - (Creative Lighting)

Unfortunately I was sick for an entire week and have been set back with workload. Due to this, I also have missed out in getting feedback that I may have needed earlier.

This week I splayed specifically for adding creative lights. The last few weeks I had been building up a knowledge about getting more control in Maya whilst also keeping with the quality/appearance of a day lit scene.

METHOD

Aim

The main thing I wanted to concentrate on was to develop a lighting system that was readable and enhance the scene. Naturally I wanted the center court to be working as the focal point as that was the designated stage for action. I also wanted there to be a sense of depth in the shot, so the viewer wouldn’t feel closed in when viewing.   

Fig.1, Base image lit

Setup

In order to establish a base lit scene to work with I used the previous lit scene and added to it. As a further update for the actual environment I chose to have more assets in there to give a more motivation for creative light.

Fig.2, Older light rig

Creative lights added:

-       -   Area lights to light up characters faces  

-       -   Target light to brighten up the center of the court

-        -  Coloured lights for the subjects backdrop (stylistic)

-        -  Golden ambient light to warm up the shot

-        -  Another area light for asset

Fig.3, Light setup from afar

Fig.4, Light setup close

As an addition to creative lighting I also began to remove various lights to supplement for the expense in render time. All the bounced lights were removed and replaced with Final Gather as it was easier to manage and less time to render.

Lights removed or changed:

-        -  Hot distant area lights removed

-        -  All bounced lights (lighting stalls, walls and lamps)

-        -  Distant light increased greatly to light up hill in the distance

-        -  Sky area light replaced with ambient light and light linked solely to it

-        -  Directional light intensity reduced

[image]

RESULTS

Fig.5, Final result

For what I had created I took reference from the environmental lighting in ‘One Man Band’ as it had a similar hierarchy in accordance to what I had.

Fig.6, One Man Band (Andrews & Jimenez, 2005)

AFTERMATH

Light Linking

Another issue I discovered unexpectedly was the practicality of link linking in a 3D animation. As far as setting it up in a scene it is an effective and necessary tool that gives more control over what receives light. When the light rig is exported and referenced into a different scene it seems to lose the retained information and reset all of the specific light connections. The reason to this seems obvious though. If you are importing a rig the information would be retained ideally in connection with the name of the asset and possibly the kind of object itself. When the lights enter a new shot they do not recognise the objects they are meant to connect with and therefore do nothing (aka reset).

I haven’t exactly solved this problem but have promising strategies in dealing with it.

This include:

-          Mel scripting

-          Asset manager

-          Easy Light Link

This wasn’t so much something that I didn’t necessarily need to figure out, but came out of my research.  

Final weeks

Recently I actually did get feedback (despite the major delay) after finally showing the results of my progress. What entailed was a critical kick back into reality of what was needed to be achieved. My thesis was not so much aesthetic based but rather efficiency based relative with the best quality possible. I haven’t really been focusing on the actual timing of my renders which was the point of it all.

My new goal is now clearer. Focusing on trying to get an equally appealing lighting system that is more efficient than using Physical sun and sky. This won’t necessarily mean that I’ll be diverting from other renderers but story telling light has to be axed.

Learning Contract

Updating from new goal

Week 10 – Working out possible ways to lower render times whilst keeping good quality in Maya. I also may show the advantages to having more control over scene lighting rather than Physical sun and sky pre-sets.

Week 11 – Looking at the effect of different renderers within Maya and how they may speed up render time.  

References

Andrews, M.(Director), & Jimenez, A.(Director).(2005).One Man Band [Motion picture].CA, Pixar.

Week 8 - (Lighting and Rendering)

This would be the third week that I have been working towards creating a believable daylight scene. As much as it is frustrating that I have been on this for so long it is good that I have a much more thorough understanding of lighting in Maya.

Anyway, advice I got last week was to look into 'Lighting and Rendering' as well as HDRI lighting and colour management. I ended up actually borrowing the book and was wondering why I didn't find it before. A lot of it covered stuff I had already knew (had previously researched) but still came in handy to read.

Method

For the actually experiment for this week I was going to finally study photography and try to simulate a similar lighting setup as I could. Though before I could do that I had to remove the exposure node that I didn't want to be dependent on. From what I learnt, the way to effect exposure without using the node is through colour management.

Fig 1. Base from week 7

Originally all I knew about colour management was at the top of rendering settings. It also lies within each texture that is imported and through the render screen (was hard to find). Through there the image profile is set to linear sRGB and the Display is set to sRGB. This is meant to simulate the correct lighting for most computer screens apparently. This was rendering as an .EXR in order to adjust the gamma correctly (2.2).

Fig 2. Exposure adjustment manually

In addition the light rig I set up other lights to accommodate for bounced light. This was mainly target lights around the center  court as it was the location where the eye would look. There were also area lights put in the back to give variation to the grass (though the shadows do disappear a bit).

Fig 3. Light Setup 1 

Fig 4. Light Setup 2

Fig 5. Final 

HDRI lighting was played with a little. The method which i think was correct that I took was adding it as a skybox then  Final gather it across the scene. The results looked good but the dispersion of light got rid of my hard shadows (maybe it might look good for distant stuff). It also seemed like a mental ray dependant system (of what I discovered). Technically I still use mental ray, but I know I can easier change to other renders with little tweaks rather than have build in mental ray stuff to my rigging.

Fig 6. HDRI (sadly not a good example)

Learning Contract

Thanks to illness and being stuck in IND my learning contract has not only gotten behind but is in need of an update.

Essentially I have covered enough in order to move onto  creative lighting and other things I said I would do.

For week 9 I'd like to move onto creative lighting setups and manipulating the scene that I am currently using. Playing with temperatures of light will be introduced here as well as I have missed out on that week.

For week 10 I'll be looking at the effects still of a couple of renderings in order to both support my research to making efficient light rigs and use of average consumer hardware.

Week 11 might all spill into week 12 but will stand as the amassing of all the work I have done on my experiments. It will be fixated towards trying to create an emotional experience for the viewer through lighting. This will also may be altered through different angles of the scene to have a more depth look into the lit environment. There is a chance this may not happen due to deviating from the thesis but would be cool to end on. I may ask just to clarify.   

References

Birn, J.(2010). [digital] Lighting and Rendering. (2rd edition).Berkeley, CA:New Riders

Week 7 - (Light Rig Base Part 2)

Last experiment was a bust

Not only did I panic and use a pre-set mode that worked but I was unable to customise it to my likeliness. The key here is that it worked.

This week my goal was again to achieve a default lighting system for a 3D animation. This time I would be breaking the network of physical sun and sky down (rather than wing it). I had a hunch that if I could understand the parts of how the mode was built, I could customise it to my likings.  

Physical Sun and Sky (PSS)

The ideal lighting situation I was aiming for was an average sunny day in a park at about noon. 

Rather that delve into the web and look into how (PSS) is built, I started by starting with a directional light a moving on from there. The directional light seemed somewhat ordeal due to experiments  I had done in week 4. Bah, Enough tangenting.

In PSS the Directional light has suppressed custom shaders. In both Light Shader and Photon Emitter Physical Sun has been placed. By adding custom shaders both the intensity and shadows features become locked and arent able to be manipulated through the light anymore (my problem with the last experiment). 

Within the Physical sun is the only place to manipulate the shadows and that's just how soft they are.

Physical Sun has further connections with physical sky, connecting most features for them to work well together.

The time of day gets controlled by having the directional light (or its world attributes) connection through the sun slot. This allows the environment to change from a midday setting to a golden evening. 

In order to see what is on screen effectively custom shaders are connected to the camera. This being Physical Sky in the environment shader to view the blue sky and Exposure in the Lens shader.

Finally to help support the distribution of a one light scene final gathering is turned on.

Altogether this builds the product of PSS that is a usable system that can make an environment look good as well as its time of day. For me though it wasn't good enough.     

    

What I did

After building this from scratch I started to rip out stuff that I felt was hindering me. 

- First to go was the Physical Sky. Having that in the connected to anything is what was responsible for the bluish tint I hated. Stripping it out of the camera shaders did that. 

- There was also no need to have Sun Direction as I was sticking with midday so that was stripped out. 

- I wanted to be able to manipulate shadow opacity and light intensity so the Physical Sun was taken out  (as well as Sky then). 

- Turning off the final gathering brought back thick black shadows. Adding 2 Fill lights for the side and front lit them up. Adding a yellow ambient light goldened up the scene to simulate sun light affecting the environment.

 - Adding a light for the distant background only added depth to the scene to simulate light hitting distance.

- A final light was added strictly for the skybox to the light up a sky. Mainly because the Directional light could not work properly within a dome. 

Results

 Before

 

 After

 

Final thoughts

The results produced default lighting I was comfortable with. From here there still needs to be creative lighting added in order to really help this scene come alive. I just glad this is all starting to work now.

Week 6 - (Light Rig Base)

For this week my goal was to actually produce an initial light rig. This rig would be able to simulate morning, midday, evening and night. From what I had known about light already in Maya and in real life I had originally seen this an average task for the time I had. The problem was that I had just assumed certain things could be done when I hadn't fully experimented with them at all. This being shadows.

I started by drawing out some schematic light diagrams for what I assumed may have been a good plan. It worked out in theory and was able to translate as expected. It wasn't till I tried experimenting with Physical sun and sky did things start to go sour.

PHYSICAL SUN AND SKY

The main reason I used this is that it had already useful attributes for environmental lighting. Through this I would adjust things accordingly as well as tweaking colours and adding extra lights. This would work fine in theory but when I tried it I hit a brick wall.

I have NO idea how to manipulate shadows.

Changing the time of day and the actual softness were the only ways I figured out how you could manipulate shadows. This was bad very bad. After days of attempts and playing with everything I still couldn't figure this out. It was important that I did as I was getting light bluey shadows that had physical sun and sky written all over it. Without the ability to control shadows properly I couldn't continue. I stopped the experiment and went with failure.

(sigh)

WHAT I LEANT AND WHERE TO GO FROM HERE

As far as my skill set is right now I feel that I unfortunately have to call a bust on physical sun and sky. It’s probably better too as being able to create my own light rig manually allows for more control and is still usable if I chose to switch to different renderers.

Directions I could take for future experiments may involve working with lights individually. I may also be directed to see the impact of how different shaders/mip nodes may influence. Could a shader have fixed my shadow problem?

For next week I may have to alter the learning contract to accommodate for the fact that I haven’t finished an actual rig. Combining weeks together (like seasonal and create light weeks) may bring me back to where I need.         

Week 5 - (Fundamentals)

To recap on last week, I was testing out different light sources of Maya and experimenting with their attributes. From it I was able to get a basic understanding of how different lights work in Maya. 

This week’s focus was on the actual theory behind lighting and light rigging itself. Rather than just focusing on CGI lighting I chose to look into other forms of media. This included; film, photography and even old paintings as lighting had been a thoroughly researched topic in these fields. Not to mention they were using actual light that gave natural results (rather than just programed).

The actual challenge is trying to create believable lighting within the capabilities of Maya. It much simpler to have bad lighting in a computer program than it is real life. Some things can easily just be seen as ‘out of place’, ‘too bright’ etc. A knowledge of how real lighting works can be transferred to Maya and give believable results that are replicated in real life. There are advantages and disadvantages through the program but for now my focus is on observation. 

Factors of good lighting

"Good lighting is important for quality video in three different ways: exposure. Illusion of depth and mood/feeling", (Jackman, 2010)

This was probably the more basic way of summing up production lighting that I understood. The three factors it address’s (exposure, illusion of depth and mood/feeling) are all needed when setting up shots.

Exposure is the visibility of a shot. How lit the overall areas of the shot are and what details and forms are shown. Having not enough exposure on a shot causes dark hard to read visuals, whilst over exposure blows out whiteness. 

Illusion of depth is simulating 3D space within an environment. Even though a camera can capture real life, the actual viewing of the material is only in a 2D surface. Having depth in a scene tricks the viewer into thinking (or rather reinforcing) there is depth in the shot. 

Mood and feeling help give a soul/deeper meaning to the shot. This could be highlighting something or telling us something about the character/object. Having this brings a shot out of realism and more into creative, artistic and semiotic communication.   

Lighting in traditional work  

To start initial research on light I looked into old paintings that had a direct demonstration in the subject of light. It was an interesting direction to take the research as light was being understood in art before film and photography. The Painting that I ended up looking at was Claude Monet's series of haystacks. From looking at the pictures there is a rapid change in the tone and colour. The haystacks are basically the same but painted at different times of day and weather. What can be learned from this is the dramatic changes natural lighting can have on objects. How the different lighting can give off alternate moods and behaviors.  

Lighting Rigs

High key lighting

High key lighting is a two light set-up of both a key light and a fill light. They both are angled in the front of the shot with the fill light having near the same intensity as the key. This results in everything being evenly lit and at a higher intensity. With the extra dimension of colour forms are separated and readable (unless if you are colour blind). The advantage behind this system is that it is simple and it works. The negative is that it creates a wash of light that makes shots look flat and is bland.  

3 point light system

Out of all the research I was able to do for this week the 3 point light system seemed to be the most commonly written about light rig. Though you couldn't really blame it considering of how much of a effective staple it is to lighting. It order to produce the 3 point light system you place a light angled on one side of the camera focusing on the asset/area. This is the dominate light source and is known as the key light. On the other side of the camera is another light but with much less intensity. The light, known as the fill light, exposes the areas which the key light would leave in darkness. Having the light at less intensity creates light tone and contrast that helps bring out details. Finally the is a light placed at the back of the asset/area. With the light, or rim light, protruding off, the asset/area creates a highlight that helps give a sense of depth within the shot.  

Light bounce

This isn't so much of a rig as it is a technique. When we have an existing 3 point light system an extra key and fill light can be added to control bounced light on an object. The secondary lights will be in line vertically with their duplicates and have greatly lowered intensity. This adds brightness to the lower part of an object simulating light that has bounced from the ground. A slight change but a definitely noticeable adjustment.

Building light plans

When it comes to lighting for environments it's a good idea to produce a draw plan first of where the lights may go/where you need lighting. 

Unfortunately I couldn't exactly find any kind of formulaic environment light rig. However with the varied location choices it would only seem silly for there to be a one all rig for environments. I did however find a process in developing them. 

 Primary light source - Add lights and attributes of those lights that would come naturally to the shot being created. This would be though sunlight or artificial light, like lamps.  

Lights focused on active areas - Giving extra or special consideration for lighting where the actual action may take place. If the camera is focusing on a character, they should have more lighting focus to them. 

Lights to expose and give depth - Creating a focal point to expose  then adjust the rest of the shot according. 

Set lights properly - Basically as it says. placing the lights that have been planned.

Watch shadows - Have control over the shadows. Make sure no odd shadows are intersecting weirdly with the action of the shot. A shadow is a form.

Accent lighting, interesting light structure - Breaking up spaces with lights and shadows to create more interesting visuals.

Fill lights - Filling up all the darkness that would be otherwise caused by the original key lights.  

Shadows giving sense of depth - Manipulating shadows whether it be adjusting for more realism or creative lighting.

Eliminate hot spots - Areas that may have a fairly strong lit area that are out of place in the shot. The previous steps reduce that happening. However if there is movement in the scene the lights may need to be adjusted accordingly.

This is obviously more focused towards film production , however the techniques themselves transfer directly to CGI.

Other useful stuff I discovered whilst researching light

Barn doors - controls aperture of light

Mirror ball - shows overall lighting of scene

Probably finishing this on a rough note but for some reason I writing too much from what I have intended. There has just been so much information to learn. I could probably adjust the scope of the research but I still really want to focus on efficient environmental light rigging.

References

Jackman, J.(2010). Lighting for Digital Video and Television. (3rd edition).Oxford, UK: Focal Press

Parrish, David A.(2002). Inspired 3D lighting and compositing. (1st edition).Great Britain : Premier Press

Unknown. (2013). Turner to Monet. [Online] Available

http://nga.gov.au/exhibition/turnertomonet/Detail.cfm?IRN=29073 [October 22nd 2013]

Images

Ashton, K. (2011). Studio Lighting Techniques. [Online] Available

http://www.cse.ohio-state.edu/~parent/classes/682/WI12/TechReports/lighting.pdf [October 27th 2011]

Ball, A. (Producer), & Mendes, S. (Director).(1999). American Beauty [Motion picture]. American: Dreamworks.

Feringa, B. (2013). Bert Feringa Blog. [Online] Available

http://feringa0.home.xs4all.nl/weblog/hay-3-claude-monet.html [March 29th 2013]

Unknown. (2012). Light Your Video. [Online] Available

http://www.howto.gov/social-media/video/light-your-video [May 4th 2012]