Thursday, September 1, 2011

The Eagle Has Flown


Home, finally.

It's been the longest and best 3 months of my life.

I'm kinda starting to miss L.A. now actually.

Well what I've accomplished during this trip has shown me that anything is possible.

Going to L.A. has made me realize that there's so much more to life outside of Asia.

It's just so... Awesome.

I really don't know how to explain it but the moment I cruised around my neighbourhood with my skateboard earlier, it felt dull unlike America.

Heheh, maybe I just need to get used to the conservative asian surroundings.

So here it is, part my life as a VFX Artist and my lifelong dream to be a movie director.

An attachment to Side Effects Software in L.A.

Success.

-

WAIT!

Who said this is the end of my adventures?

In fact, this is just the beginning of my adventures as a VFX Artist!

I'm just taking a short break, chill!

Tuesday, August 30, 2011

Day 57

Countdown To Departure: 0 day(s)

Tonight is the night we leave L.A. as Champions.

Monday, August 29, 2011

Day 56

Countdown to Departure: 1 day


HAPPY BIRTHDAY MICHAEL JACKSON!


I grew up with your songs and I'll never stop listening.


R.I.P


-


We had one last meal at Denny's for lunch today.

I bought a new Tokidoki cap for myself since it kinda fits my Tokidoki skateboard deck.

Time feels really really slow while we wait for our flight back home to Singapore.

Sunday, August 28, 2011

Weekend 12


Countdown To Departure: 2 days

12 weeks have officially passed as predicted in our IAP schedule.

We went out with Steven and Shih Peng to Cheesecake factory for lunch today.

The food was AWESOME.

The cheesecake however was AWESOMER!!!

I know there's no such word as awesomer.

However I typed it out so it counts as a word right?

Friday, August 26, 2011

Day 55

Just for the sake of post consistency I shall post something for today.


Nothing happened.


Just kidding, we went back to SESI again to give our colleagues chocolates since every single one of them have been VERY nice to us.

Other than that if you've just visited this blog, please help yourself to viewing my final output of this project in the previous post before this.

Cheers!

Thursday, August 25, 2011

The Presentation

Day 54

The presentation was a success!

I spoke with confidence and I had fun presenting to everyone at the office.

Within time limit of 10 minutes, short, sweet and awesome.

Here are the final outputs:

Project Hellfire: Side Effects Software Attachment Project
Artist: Bryan Isaac Emmanuel
Location: Side Effects Software L.A.
Softwares: Houdini, Nuke, Photoshop
Production Duration: 2.5 months
Deadline: 25th August 2011

Project Completion: 87.5%

Improvements will be made when I feel like working again.

Link: http://www.youtube.com/watch?v=FYs_BuitEoc


Shot Breakdown
Link: http://www.youtube.com/watch?v=ok7Gyy_o4yI



Apache Turntable
Link: http://www.youtube.com/watch?v=dEmulFFKmdc


Convoy Turntable
Link: http://www.youtube.com/watch?v=MuZZsI9XhSo


Thanks to the people who made this from impossible to possible:

Mentor
Steven Ong

Side Effects Supervisors
Aliza Sorotzkin
Ari Danesh
Ziggy Ostertag

Special Thanks
Charles Trippe
Dan Bodenstein
Eric Araujo
Evan Boucher
Jia Bao
Lu Sheng
Mohamed Ziyaad
Rap Cheng
Tanner Chow
Zack Zhao

-

Since last year December we've been working non-stop since our year 3 modules started along with Studio Project, Final Year Project and Side Effects Software Attachment Project SESI with no breaks at all.

Nine and a half months of non-stop CG action creating visual effects and animation.

Not even a single holiday.

Well guess what,

Our flight back to Singapore is next Tuesday.

Our holiday period finally starts next Wednesday!

Wednesday, August 24, 2011

Day 53

Countdown to Final Presentation: 1 day

1) Prepare presentation slides. [in progress]

2) Sound FX [done]

3) Set up and render turn tables. [done]

4) Composite passes breakdown demo. [done]

5) Final MasterBeauty Composition Render. [in progress]

-

I'm re-rendering the terrain pass because I just realized it had swimming textures.

Yes I may sound calm right now but actually I'm freaking out.

-

I'll show whatever I have tomorrow.

I'll improve on project Hellfire when I return to Singapore?

Tuesday, August 23, 2011

Day 52

Countdown to Final Presentation: 2 days

1) Prepare presentation slides. [in progress]

2) Sound FX [in progress]

3) Set up and render turn tables. [in progress]

4) Composite passes breakdown demo. [in progress]

5) Final MasterBeauty Composition Render. [in progress]

Just a few more frames to render for my explosions.

Today we're going to visit Sony Image Works at 2pm.

-

Master Beauty Test Render


Testing out edge-blurring of explosion footage in progress.

Monday, August 22, 2011

Day 51

Countdown to Final Presentation: 3 days

Here's a recap of what I must complete by this Wednesday:

1) Prepare presentation slides. [in progress]

2) Sound FX [in progress]

3) Set up and render turn tables. [in progress]

4) Composite passes breakdown demo. [pending]

-

1:35 pm




Getting my turn tables ready.

Sunday, August 21, 2011

Weekend 11


Announcement: Stay tuned for the finalized short film "Operation Hellfire" which I will upload onto this blog on Friday!


Till then I won't upload much pictures which might act as spoilers.


We headed back to the office yesterday and today thanks to Eric for letting us in with his key.

I've checked my renders and all that's left to render are a couple of frames of the main explosions.

An extra element that I wish to re-render tomorrow is the fog.

That's about it.

4 things I have to complete by this coming Wednesday.

1) Prepare presentation slides. [in progress]

2) Sound FX [in progress]

3) Set up and render turn tables. [pending]

4) Composite passes breakdown demo. [pending]

Friday, August 19, 2011

Day 50

This project is giving me the goosebumps of epicness as the end draws near.

This is the final run.

Here's a list of my rendering progress:

Particles

Missile Trails (Render Farm: 99%) - Everything rendering according to plan.

Convoy Dust (Render Farm: 99%) - Slight hiccup as opacity changes at mid point, tried to fix twice but didn't work. Can be fixed by just rendering a couple of frames and mixing two layers of dust in nuke.

Damaged Convoy Smoke (Render Farm: pending) - Still deciding whether to send remaining 98 frames to render farm to render due to long queue.

Machine Gun System (Workstation 12: pending) - Frames remaining, 92.

Simulations

Explosions 01, 02, 03, 04 (Render Farm: 45%) - Elapsed Time 3 days 23 hours. ETA 3 days 20 hours.

Explosion 05 (Workstation 12: 97%) - Rendering 1 frame left, frame 650 which was screwed up.

Explosion 06 (Workstation 07: 25%) - 31 frames left.

Shadows

Setting up a new file for rendering remaining explosion shadows and zdepth pass.

-

11:50 am

SCREW UP ALERT! SCREW UP ALERT!

Just found this out when compositing the missile trails and convoy dust particles in nuke.

Missile trail particles and convoy dust particles have been screwed up due to wrong intensity.

However if I decrease the intensity I won't receive ANY particles.

Particles

Missile Trails -Setting up

Convoy Dust - Setting up

Damaged Convoy Smoke (Workstation 08: 1%) - Just sent to Workstation 08

Machine Gun System (Workstation 12: pending) - Frames remaining, 92.

Simulations

Explosions 01, 02, 03, 04 (Render Farm: 45%) - Elapsed Time 3 days 23 hours. ETA 3 days 20 hours.

Explosion 05 (Workstation 12: 75%) - Rendering 1 frame left, frame 650 which was screwed up.

Explosion 06 (Workstation 07: 27%) - 31 frames left.

Shadows

Setting up a new file for rendering remaining explosion shadows and zdepth pass.

-

3:25 pm

Particles

Missile Trails & Convoy Dust (Workstation 12: 25%) - Everything rendering according to plan.

Damaged Convoy Smoke (Workstation 08: 80%) -Rendering according to plan.

Machine Gun System (Workstation 08: pending) -Waiting for damaged convoy smoke to finish rendering.

Simulations

Explosions 01, 02, 03, 04 (Render Farm: 46%) - Elapsed Time 4 days 03 hours. ETA 4 days 22 hours.

Explosion 05 (Workstation 12: 100%) - done

Explosion 06 (Workstation 07: 25%) - 31 frames left.

Shadows

Setting up a new file for rendering remaining explosion shadows and zdepth pass.

-

3:46 pm

Particles

Missile Trails & Convoy Dust (Workstation 12: 28%) - Rendering according to plan.

Damaged Convoy Smoke (Workstation 08: 100%) - done

Machine Gun System (Workstation 08: 15%) - Rendering according to plan

Simulations

Explosions 01, 02, 03, 04 (Render Farm: 46%) - Elapsed Time 4 days 03 hours. ETA 4 days 22 hours.

Explosion 05 (Workstation 12: 100%) - done

Explosion 06 (Workstation 07: 30%) - 27 frames left.

Shadows

Setting up a new file for rendering remaining explosion shadows and zdepth pass.

-

5:58 pm


Alright things are a little less hectic around here and I've calmed down.

I've fixed up my composited scene and solved several issues in nuke.

Particles

Missile Trails & Convoy Dust (Workstation 12: 50%) - Rendering according to plan.

Damaged Convoy Smoke (Workstation 08: 100%) - done

Machine Gun System (Workstation 08: 100%) - done

Simulations

Explosions 01, 02, 03, 04 (Render Farm: 47%) - Elapsed Time 4 days 06 hours. ETA 5 days 12 hours.

Explosion 05 (Workstation 12: 100%) - done

Explosion 06 (Workstation 07: 40%) - 26 frames left.

Shadows

Rendering beauty.EXR with Direct Shadow Matte Pass (Workstation 08: 2%) - 147 frames left.

Thursday, August 18, 2011

Day 49

Updates:
1) Mini explosions for destruction of last 2 tanks. [done : monday]
2) Finalize machine gun particle system, including sprites. (Requesting help from Ziggy) [done : tuesday]
3) Fix missile trail rendering problem. [done : monday]
4) Re-render all particle simulations. [in progress : thursday]
5) Re-render shadow matte pass for whole scene. [in progress : thursday]
6) Re-render fog & ambient occlusion passes. [done : wednesday]
7) Composite rendered shadow matte, fog and ambient occlusion passes. [done : wednesday]
8) Colour correct scene in comp. [in progress : thursday]
9) Check and finalize all simulations. [done : monday]
10) Batch render all remaining simulations. [in progress : thursday]
11) Batch render zdepth [pending]

An "Intro to Houdini" class was conducted at 12pm today and just ended.

So I had to interrupt the renders on several workstations.

Now the renderings are back online and rolling.

Right now I'm still documenting my stuff and getting ready for the presentation.

Everything is going according to plan. Yeah I know I tell myself that in this blog all the time.

Why? Positive attitude man. Positive, positive, positive!

Without it nobody's gonna go home with any accomplishment.

Mr Douglas just sent in an email giving us yet another motivational push.

Thank you all who have been supporting us since Final Year Project started, followed by Project SESI.

It's a race against time and I know I can do it.

We'll all come home as champions.

Wednesday, August 17, 2011

Day 48

Today we received an email from our best ol' friend back in Singapore, Rap.

Thanks to him I've regained one last burst of motivational energy to finish Project Hellfire with success.

Updates:

1) Mini explosions for destruction of last 2 tanks. [done : monday]

2) Finalize machine gun particle system, including sprites. (Requesting help from Ziggy) [done : tuesday]

3) Fix missile trail rendering problem. [done : monday]

4) Re-render all particle simulations. [in progress : wednesday]

5) Re-render shadow matte pass for whole scene. [in progress : wednesday]

6) Re-render fog & ambient occlusion passes. [done : wednesday]

7) Composite rendered shadow matte, fog and ambient occlusion passes. [done : wednesday]

8) Colour correct scene in comp. [in progress : wednesday]

9) Check and finalize all simulations. [done : monday]

10) Batch render all remaining simulations. [in progress : wednesday]

11) Batch render zdepth [pending]

When I came into the office and checked the completed renders, I was like half YAY and half WTF.

My ambient occlusion and fog passes have successfully completed rendering.

That's the yay part.

The WTF part however...

My 2 small scale explosions failed to render in the render farm and the image sequence showed nothing.

BUT

I've setup the explosions to render on a machine this time and it's working out pretty well so everything is back on track.

All that's left to do now is to maintain order and stability in my rendering files and composite the completed renders that consistently come in.

At the same time, I'm documenting all my files, finding sound fx and preparing for the final presentation.

-

5:34 pm

Dan taught us the basics of using nuke today.

I'm still documenting my files.

The explosions are still being rendered in the farm at 25%.

The missile trail and convoy dust particles too at 90%.

I've been making sure the renders are stable and they are so far.

Tuesday, August 16, 2011

Day 47

Here's a recap of what I'll have to accomplish by tomorrow.

1) Mini explosions for destruction of last 2 tanks. [done : monday]

2) Finalize machine gun particle system, including sprites. (Requesting help from Ziggy) [done : tuesday]

3) Fix missile trail rendering problem. [done : monday]

4) Re-render all particle simulations. [in progress : tuesday]

5) Re-render shadow matte pass for whole scene. [in progress : tuesday]

6) Re-render fog & ambient occlusion passes. [in progress : tuesday]

7) Composite rendered shadow matte, fog and ambient occlusion passes. [pending]

8) Colour correct scene in comp. [in progress : tuesday]

9) Check and finalize all simulations. [done : monday]

10) Batch render all remaining simulations. [in progress : tuesday]

-

11:44 am

I've just realized that once I finalize my machine gun particle system, I'm only left with rendering all my remaining stuff, sound fx and compositing!

Oh but that doesn't mean I can slack off and not improve my scene...

but well... what the heck

YEEEHAAAAAW!

Let's just pray that the renderings will be fast as I've put them into the render farm.

-

3:09 pm


I've finalized the machine gun particle simulation, bullet casing drops AND smoke emission from the damaged convoy.

We've just received an email from Mr Ari stating that our presentation is next thursday and we must get everything ready by then.

Date: Thursday, August 25th.
Time: 10:00 am
1. Duration of each presentation no longer then 10 minutes.
2. Each student display their final rendered shot.
3. Show break downs of the shot.
4. Walk us through your networks and explain what were your technical challenges and how you solved them.

Now, time to setup the last few mantra nodes for batchrendering the remaining particle systems.

-

3:37 pm


Done

Alright, now it's time to make sure that everything is rendering accordingly.

This has been the longest 3 months in my life, the most worth it too.

Let's finish this.

Monday, August 15, 2011

Day 46

My simulations are somewhat done batch rendering.

However there are a couple of errors in some frames that I'll need to fix by re-rendering.

Here's a list of stuff I aim to accomplish by Wednesday.

1) Mini explosions for destruction of last 2 tanks. [done : monday]

2) Finalize machine gun particle system, including sprites. (Requesting help from Ziggy)

3) Fix missile trail rendering problem. [done : monday]

4) Re-render all particle simulations. [in progress : monday]

4) Re-render shadow matte pass for whole scene. [in progress : monday]

5) Composite rendered fog and ambient occlusion.

6) Colour correct scene in comp. [in progress : monday]

7) Check and finalize all simulations. [done : monday]

8) Batch render all remaining simulations. [in progress : monday]

-

4:28 pm


I've put in the last two smaller scale explosions for the convoy victims of the machine gun spray.


As for the composition, I have the temptation to follow Gears of War lighting style closely.

-

I've just realized why my missile trails were not showing up.



I had to bump up my keylight intensity to 250 so that the particles would finally show themselves.

Things are going according to plan and it better stay that way.

The end of Project Hellfire is near.
Weekend 10

Post reserved for 4th visitation to Hollywood.

Friday, August 12, 2011

Day 45

So we're back in the office.

After a week of crazy adventures in Vegas, Grand Canyon, Beverly Hills and Knott's Berry Farm.

I'll post my trip experiences in the post below when I have time over the weekends.

For now let's get to work.

My pyro caches are 90% complete. Just a little bit more rendering and I'll be on my way to working on the pyro shaders.

Things to accomplish today:

1) Finalize particle system for machine gun fire and shell casings.

2) Finalize pyro simulations in lighted scene.

3) Fix terrain fog.

4) Test pyro shaders.

-

7:08 pm

I've got a couple of issues in which I need to approach Ziggy for solutions next week before I can pronounce it finalized.

The fog has also been fixed thanks to Steven. I've finally managed to batch render it out.

All pyro explosion simulations in the scene has been cached out and I'm leaving them to batch render over the weekends. The shaders seem fine at their defaults. I'll tweak them if needed.

Also I'm leaving a workstation to re-render the ambient occlusion for the scene because the previous render screwed up by giving me 22 missing frames.

Screenshots and a latest flipbook sequence will be uploaded in this post soon.

-

12:12 am


Project Hellfire: Final Scene Lighting/Simulation Flipbook Test 01
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=J5l9e2rQPSk




I've planned out a list of what I have to accomplish by next wednesday:

1) Mini explosions for destruction of last 2 tanks.

2) Finalize machine gun particle system, including sprites. (Requesting help from Ziggy)

3) Fix missile trail rendering problem.

4) Re-render shadow matte pass for whole scene.

5) Composite rendered fog and ambient occlusion.

6) Check and finalize all simulations.

7) Batch render all remaining simulations.

It's gonna be a hell of a week.

Well since this is project hellfire, why the hell not?

LET'S FINISH THIS!
Post reserved for trip to Las Vegas, Grand Canyon, Beverly Hills, Hollywood and Knott's Berry Farm.

Friday, August 5, 2011

Day 40

Today's plan is to finally get the whole scene finalized with simulations and particles inside.

I didn't get the chance to explain what I did in yesterday's post so I'll explain it in today's post.

Oh and we're leaving for vegas tomorrow.

I'm feeling pumped and excited.

-

1:09 pm

Update


Machine gun fire particles emission enabled.

-

Saturday
1:11 am

I've just finished packing up my stuff for vegas and grand canyon.

We're leaving early in the morning for Universal Studios.

I'm feeling damn excited.

Here are two videos showing some new elements I've added into the scene.

1) Pyro Simulated Explosions

Note: I cached out a few frames of each simulations for testing. I've left 2 workstations to rop out the bgeo sequence of all 4 pyro simulations over the weekend. I'll see to it once I'm back from Vegas.

Project Hellfire: Simulation/ Scene
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=tsq5bpJLFSI


2) Particle Simulated/ Driven Bullets Impact

Note: The bullet particle system is still in construction. No finalization yet till next week.

Project Hellfire: Particle Collision/ Scene
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=OX1oFGmow-0


Thanks to Mr Ari for lending his account over to me for the basics bullets tutorial. I had to resort to some improvisations and upgrades for it to work in this scene.

Alright I'm gonna take a nap now.

VEGAS HERE I COME!!!

Thursday, August 4, 2011

Day 39

I'm gonna check out the simulation file Steven provided now.

-

6:52 pm


Leaving the office now, I'll talk about what I've done today when I reach home.

Wednesday, August 3, 2011

Day 38

A couple of things to do today:

1) Composite my render passes.

2) Set up ambient occlusion pass.

3) Test offset timing of simulation in scene.

4) Ask Ziggy for guidance in machine gun shooting particles.

-

3:01 pm

I'm currently compositing my passes in nuke.

Still trying to get used to the software, mostly resorting to trial and errors.

Things are going pretty well so far.

I've managed to finally get the passes composited together for each object.



I'm still setting up the ambient occlusion scene for render.




All other objects can be rendered separately with ambient occlusion.

Both apaches however are being as stubborn as hell.

Their ambient occlusion passes refuse to be rendered out for some reason.

So if there's no other solution I'll resort to rendering out the whole scene with all objects together for one ambient occlusion pass.

-

5:42 pm



I've sent the scene to the renderfarm to render out an object casted shadow layer and an ambient occlusion layer.

In the meantime, I've finally setup my compositing nuke file with all the render passes nicely in place.

Thanks to Zack for guiding me in learning nuke basics.


I've received an email from Steven which consisted of a solution to my simulation offset timing problem.

I'll check out the hip file he did tomorrow. Also, I'll proceed to doing the machine gun particles with a tutorial Ari agreed to provide using his account.

Tuesday, August 2, 2011

Day 37

Let's get back into action.

I've given myself sudden motivation by setting a goal and a reward.

After the success of Project Hellfire, I'll look forward to hanging out with all my childhood friends as much as I want back in Singapore.

For now, let's finish what I started.

Ziggy left for his meetings and said he'll be back to review my particle simulations.

So while waiting I'll do a couple of pyro test renders using the render farm and make some improvements to the particles myself before Ziggy comes back.

This is one of the best references I've found so far:

Apache Rocket Show
http://www.youtube.com/watch?v=i2zx502RHCA

I'm still thinking if it's a good idea to over exaggerate the explosions from the hellfire missiles.

For the sake of awesome visual effects hmm?

-

4:01 pm

Ziggy gave me some cool tips and guidance to doing a missile trail.

He showed me some references from his demo reel:
http://www.youtube.com/watch?v=n0soc8xV-WA

Ziggy mentioned that when using sprites and particles to simulate smoke, always control the simulation using color, alpha and scale ramps.

Here's an important factor Ziggy asked me to take note of when working with sprites/ particles.

Sprite Alpha: Starts Transparent => Almost Opaque => Translucent => Transparent

Sprite Scale: Starts 0 => Medium => Big => Bigger




I'm still working on it and adding improvements.

I tried the method on the convoy dust emission too.


Ziggy is busy having a conference meeting with dreamworks.

I'll ask him for more advice when he's free.

Monday, August 1, 2011

Day 36


Tweaked and improved the stone shader a little.

Tanner assisted me in adding more bump and roughness to the shader.

I've made new adjustments to the pyro secondary smoke simulation.



Re-rendering main terrain overnight in the meantime.

-

5.36 pm

I'll be honest.

My godbrother's passing is affecting me quite badly.

I can't concentrate that well today.

Give me till tomorrow.

I will be back to normal to ace project hellfire.

Sunday, July 31, 2011

Weekend 8


My Godbrother, Raphael passed away at 7.15pm singapore time yesterday...

Aunt Irene, Aunt Debra, Godma Shirley, Raphael...

Watch over me as I will not let you guys down.

I'll make you proud and I'll achieve my dream in becoming a movie director.

Thank you Raphael for our childhood memories.

I love you bro.

Friday, July 29, 2011

Day 35

I'm transferring all my project files into the render farm folder where I can finally start batch rendering my scene with separate geometries, fog, simulation effects etc.

Re-referencing of geometries and textures in progress...



Post will be updated.

Let's hope everything will turn out fine!

-

4.38 pm

I've been doing test renders using the render farm for quite a while now.

So far I've had not much issues (only a little but solved) and I should be able to leave the render farm working on my scene over the weekends.

Here's the geometry separation & render passes setup I'm using to test over the weekend.

Apache01

Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte

Apache02

Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte

Road

Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte

Convoy

Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte

Terrain Main

Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte

Terrain Background

Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte

Sky Sphere

Micropolygon

Beauty

-

Here's the official setup I'm using for final renders.

Apache01


Micro/ PBR


Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte
Ambient Occlusion

Apache02


Micro/ PBR



Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte
Ambient Occlusion

Road


Micro/ PBR


Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte
Ambient Occlusion

Convoy


Micro/ PBR


Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte
Ambient Occlusion

Terrain Main


Micro/ PBR

Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte
Ambient Occlusion

Terrain Background


Micro/ PBR


Direct Diffuse
Direct Reflect (Specular)
Indirect Reflect (Reflection)
Direct Shadow Matte
Ambient Occlusion

Sky Sphere


Micropolygon



Beauty

Fog

Micropolygon

beauty

-

I'm having problems batch rendering out fog in HD. I'll figure out ways to optimize my fog renders.

I'll add ambient occlusion to the current setup I'm using after the weekend test render.

At the same time I'll watch some nuke tutorials over the weekend since I have to consistantly keep using it to separate out the passes and test composite from the exr image files.

I've decided to use .exr format instead of .png as exr is much easier to use since it compiles all the passes together.

Nuke is a good compositing software for separating out the passes.

-

5.49 pm

Aww crap!

Jia Bao just reminded me that Ariinstructed us to only use the render farm from Tuesdays to Thursdays every week.

I wasn't even supposed to use the render farm earlier oops :P

Anyway I've let the workstation render over the weekend instead.

Let's see how it goes.

Thursday, July 28, 2011

Day 34

We've just attended another of Jeff Wagner's weekly houdini class.

He taught us about the usage of geometry fog volumes.

Here are some tests I ran.



The first screenshot was rendered using ray-traced shadows. The time taken to render it was extremely long.

The second screenshot however was rendered without any shadows casting from the fog. The time taken to render was much faster than when rendered using ray-traced shadows.

In terms of visual appeal the fog in the first screenshot seems like it has more volume and density in it. The fog in the second screenshot seems flat, however it still shows difference in distance between the three teapots.

This is what happens when I put a camera within the fog to look at an object.


I was curious whether the positioning of the camera would affect the rendered density of the fog.

Well it didn't, I decreased the smoke density in the billowy smoke volume shader to... well... decrease the smoke density.


The fog volume was created using just a simple sphere, along with an isooffset and setting the output type to Fog Volume.

After that, a billowy smoke volume shader was applied to the sphere obj.

Tweaked a little bit of noise and smoke density in the shader, Volume Filter Width in the sphere obj render settings and that's pretty much the result I got.

I'll apply some fog to my actual scene and see how it goes.

-

2.33 pm

I've done some fog tests in my scene.



First render is without fog. Rendering time extremely fast. (obviously)

Second render has fog with cloud density of 0.01. Rendering took so long I decided to stop it halfway. Thus it's kinda pixelated.

Third render has fog with cloud density of 0.005 Rendering takes just as long as second render.

I'll resume tweaking, post will be updated soon.

-

6.06 pm

Credits once again go to Zack for spending lots of time today teaching me several methods of rendering passes and separate geometries in houdini, such as adding different takes, adding extra image planes in mantra and applying a black constant shader to "unwanted" renderable objects.

Steven also taught me other methods of separating out geometries by going into the mantra objects tab and applying rules for separate object rendering.

Here's a low quality batch render just for testing the new spherical sky ramp.

Project Hellfire: Shot 02 Test Batch Render 09
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=F42HV6RofGA


I'll leave the workstation on overnight again to render out a fog layer.

Once done I'll try compositing it in with the current output I have now.

Wednesday, July 27, 2011

Day 33

I've updated the production schedule once again.

Today's To Do List:

1) Ask Zack how I should carry out generation of fog in scene.

2) Ask Ziggy what improvements can be made to my current particle system.

3) Resume watching intermediate pyro lectures.

The latest pyro batch render with the secondary particle smoke simulation couldn't be completed due to hard disk memory error.

20 frames were rendered so I don't really see any point in putting up a composited version. Thus, I'll just upload a couple of frames.

Frame 01


Frame 05





Frame 10




Frame 15




Frame 20





Damn,

no where close to my realistic referenced expectations.


-

6:16 pm



Today credits go to Zack for spending most of his time guiding me in lighting.

We spent most of the day practicing the setting up of zdepth passes and discussing final tweaks to be made to the lightings and textures.

For the atmospheric fog effect I'll add it in comp once the zdepth pass is rendered out. Same goes for the depth of field.

Zack also taught me how to render ambient occlusion in houdini.



Charles taught me how to procedurally create a ramp for my spherical sky by creating a ramp parameter in mantra surface. No lights are needed to light up the sphere, the material acts like a constant.


I just found out that the price to pay for rendering together with the spherical-sky is that I can't render using PBR. I'll render using separate mantras anyway so it shouldn't be a problem.

Tuesday, July 26, 2011

Day 32


Yeap, new trick board on the right with 52mm STF wheels . Cruising board is the one with the 60mm ATF wheels and the tokidoki pirata design on the left.

Yeap,

Double awesome.




Before I proceed to any other part of the project plan like applying atmospheric fog surrounding the terrain, setting up bullet spray particles or improving vehicle particle sprite emissions...

I shall spend today watching and analysing intermediate pyro lectures.

Like the previous intro to pyro post, I'll post notes of what I've learnt for anyone interested.

Here I go!

-

Intermediate Pyro


Anyone who has visited my previous project blog Dragon Flood might have come across me mentioning these expressions before.


Scale X Y Z
bbox("../xform_center", D_YSIZE)/3

Center X Y Z
centroid("../xform_center", D_Y) - or + bbox("../xform_center", D_YSIZE)/3

Once again credits go to Mr Ron for teaching me these expressions.

This is a post which talks about the expressions used: http://lghsh202.blogspot.com/2011/05/day-47-48-updating-in-progress.html

Well it's probably quite obvious to most artists but I'll just say it anyway. These expressions can be used to normalize and centralize bounding boxes, including fluid and gas containers.

They're useful so that when you need to change the sop path the simulation is reading data from, you don't need to adjust the size of the container.

That probably can be solved using the [gas resize field + sopsolver] method which resizes the container based on the scale of the running simulation too.

Sopgeo Node: Reads data from sop path output node. Must have data name just like gas resize field node.

Details View: Spreadsheet is good for viewing pyro data in a list. Can be used to check if SOPs data has been pulled into DOPs.

To Be Continued...

Monday, July 25, 2011

Day 31

Counting this week, I have 3 weeks left to complete Project Hellfire.

Let's start off with what I plan to do today.

1) Pyro - particles test simulations.

2) Atmospheric fog test.

3) Texturing and Shaders upgrade.

This post will be continued.

-

6:06 pm

I spent today working on the secondary smoke blast simulation using particles to drive the pyro simulation. It worked out pretty okay as a start but I'm planning to ask Mr Ziggy for more guidance regarding the motion of particles to make my simulation more realistic.

Thus I have not touched any lightings or textures in my scene yet.

I'll head home soon to composite and post the flipbook and pyro batch render tests.

-

10:11 pm

Project Hellfire: Pyro Test Batch Render 03
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=AG1cwB7-mEQ

Primary pyro simulation.



Project Hellfire: Pyro Flipbook Test Render 02
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=uYt5rUIBrIk

Secondary smoke blast test.


This is a flipbook test. Shaders have been tweaked. Scene has been left on workstation overnight to test batch render. We'll see the final outcome of this flipbook tomorrow.

Sunday, July 24, 2011

Weekend 7

On Saturday, I went over to Rip City Skate once again and got myself another skateboard customized for tricks.

On Sunday, we went to Dreamworld to shop for figurines and I got myself a nice looking catwoman model.

Other than that the rest of the weekend were spent skating to and fro grocery markets to do our mundane shopping activities.

The weekends are getting more and more boring.

At least the weekdays' office hours would compensate that, I will make sure project hellfire is a success.

Oh!

Forgot to mention, we watched Captain America and it was awesome!

Friday, July 22, 2011

Day 30

Mr Michael! I have a question for you at the end of this post!

This is another new batch render test 08 which was completed a couple of hours after the previous batch render 07.

Thus no changes were made to the camera movements yet based on Mr Michael's advice.

Project Hellfire: Shot 02 Batch Render Test 08
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=HsKpm6odXlQ


I'll let Evan take a look at this sequence and get some advice about adding human imperfection to the camera motions.

Today's production plan also consists of testing out a secondary smoke simulation for my explosion.

----

Camera Shake Solutions

Animation of camera shakes and focus

Evan reviewed my overall animation and thinks it's alright, despite of a few slight hiccups in the first apache animation as it takes off after firing the missiles.

When I brought up Mr Michael's advice to him about camera shakes and focus randomness, Evan's solution was to add a node above the camera obj and add some motion effects noise in it.

One advantage: Animating camera shakes in the scene would give it a more realistic feel than post-production since it's in 3D space.

Post-production camera tweak

Steven came over to the office to review our work in the evening.

He felt the overall animation was alright for now. Any touch ups or slight tweaks can be added towards the end just before batch rendering.

For the textures, the terrain still looks kinda plastic and he approved of me editing and painting textures for the terrain.

As for the pyro simulation I couldn't come up with satisfying results for the secondary smoke blast emitting from the explosion. Steven suggested I use particles to emit and drive the pyro simulation.

Lastly, I brought up the issue of realistic camera movements and Steven suggested rendering the frames slightly larger in resolution than what I have now. (HD720)

With that I can apply camera shakes in post-production.

One advantage: It will save me batch rendering time.

Regarding depth of field, Steven suggested rendering out a depth pass in houdini.

-

Hmm Mr Michael, both ideas have advantages and disadvantages.

What's your opinion?

Thursday, July 21, 2011

Day 29

Check out my latest batch render test.

Project Hellfire: Shot 02 Test Batch Render 07
Artist: Bryan Isaac Emmanuel
Link: http://www.youtube.com/watch?v=Qu9dCRY5UHA


I'll resume textures and lighting touch ups soon.

For now I'll take a break and test run some pyro simulations and machine gun particle effects.

----

Thanks to the pyro lectures and the notes which I've gathered, I understand what I'm doing when I go about tweaking stuff in the pyrosolver.

The previous post includes all my pyro notes I've written down yesterday.

Here are some current test renders.






These are unfinished renders as they're taking quite long to render this much already.

I'm just testing out the overall look of the simulation, no tweaking of shaders yet.

One important factor which I had to keep taking note of is the amount of time each hellfire missile strike explosion is limited to.

At the same time I'm referring to a collection of reference videos which I've posted in previous posts.

That's when I realized that I might need a secondary pyro simulation to create a volume of fierce, fast spiky smoke shooting outwards (they're flying debris actually) as seen in most explosions.

1 - 8 frames the ignition shall be the most vigorous.

9 - 25 frames smoke take over 85% of the simulation at accelerating speed.

These simulation tests are not close to perfect or realistic.

I'll have to ask Steven how he feels about my simulation test results tomorrow.