Studio 2

Snooze update

2009-04-12T23:42:00.000-07:00

"Coming to the end for this quater"

2009-03-11T22:30:00.000-07:00

Dream room

2009-03-08T23:15:00.001-07:00

New Giant Freaking Robot (website Layout) by : Brendan Fitzgerald

2009-02-26T06:55:00.001-08:00

VISIT THE SITE

"let there be light"

2009-02-24T21:00:00.002-08:00

Giant freaking robot

2009-02-24T18:50:00.000-08:00

Here are some snapshots:

Here is the link in HD

snooze red room start

2009-02-22T11:02:00.001-08:00

Eppies lighting start

2009-02-21T08:30:00.000-08:00

here is the starting point to eppies "hero lighting, and environment lighting"

character look development

2009-02-20T21:30:00.000-08:00

this is an early start still need to paint and tweak some maps especially for the body...

concept drawings

2009-02-19T14:43:00.000-08:00

drawings by joanna albright

snooze lighting update

2009-02-17T16:18:00.000-08:00

Snoozed lighting update

2009-02-07T03:38:00.000-08:00

Here is an update on the lighting with minor tweaks thanks to my fellow friends.....

Snoozed "a short by joanna albright"

2009-02-02T21:28:00.000-08:00

So I started to do some look development for Snoozed and also some light set ups still needs a lot of work and some dust in the room anyways here are some updates. Soon also eppies the skating shot will start.

72 hours and counting....

2009-02-02T08:29:00.000-08:00

Here are some new updates of the car renders, however the smoke still needs a lot of tweaking specially when it comes down to color correction and its translucency. Breakdowns will be postponed till next week to work more on the comp and make this puppy stand out yay....

by the way the new transformers trailer is out it looks promising I hope it looks as good as the first one

Room for improvements

2009-02-01T17:32:00.000-08:00

So last week we had our final critique or so we taught, anyways we came to find out that there was tons of problems with the final version like animation, smoke elements, translucency, HDRI, etc. but most of all the animation was the "key factor". The car did not look like it was moving correctly, for instance our animator made the wheels turn the wrong way, things would manually disappear "horror". Neon, Adam and I recollected ourselves after harsh all nighters to get the project back on the way. Neon went back and fix the animation, however while fixing the animation he had to use the car that had been pre-constructed to match the robot car which brought up problems for me to tackle like "shaders". Previously the shaders where updated from mia x, to mia x passes which can be used for framebuffers. Since the new car animation had to be imported the car shaders had all the old mia x shaders so I had to go back and clean up the shaders once again!!! however that wasn't a problem. The huge problem happened when I decided to incorporate a physical light into my key light attributes. The physical is great at creating an inverse square fall off, however it sucks when it decides to create it own shadow which cannot be controlled in the attributes "thanks mental ray". Also there is no documentation on how to disable the shadows in the physical light other than the tutorial from mental ray that states that by turning off raytracing the shadows go away.
Physical light tutorial "shadows off"
However we are using final gather so turning off final gather was not an option. I finally figured it out in Maya 2009 by manually disabling the shadows in the render settings at render time. Till Wednesday we will meet out deadline.

Coming to the end

2009-01-21T03:42:00.000-08:00

So Neon and I are coming to the end of this project which is due by this Friday ouch. Anyways we both encountered even more stuff that was wrong with the animation. I also sat down with Adam and figured out what was needed and what was just a waste of time regarding "render layers & frame buffers". I also fixed up the lighting which you can see here. To cut this short I have been in Monty for a very long time rendering stuff to meet our deadline. while I was rendering I saw some Iron man images on the net and it hit me maybe I should apply a bdrf Bidirectional Reflectance Distribution Function curve to almost of my shaders, and a bit of anisotropic curve from -1 to -3 so I did.

However render time did increase but not by much I also turn on the AO for all my mia_material shaders. I have a few more layers to render and I will be done and ready to hand off the images to our compositor.

-(Done)Finish frame buffers & render layers
-(Done)Paint scratches in photoshop
-(Done)Contribution pass for headlights, hood, and dgs materials
-(Done)Car paint shader flakes seperate by an RGB pass and pass contribution maps
-(DONE)Create displacement for the ground plane
-(DONE)Figure out how to get the old Mia materials to work in 2009 render passes
-(DONE)Set up new lights in the scene (hot spot)
-(DONE)Fix the shader for driver, rockets, and boosters

here are some brdf in the iron man suit, pretty neat.

Where is Waldo "Waldo meaning Pixels"

2009-01-12T00:07:00.000-08:00

So mental ray's dgs shader and dielectric shader (Diffuse, Glossy and Specular) does not seem to work with frame buffers.....ok why not may I ask, could it be because maybe those shader have been out sourced by the mia_material x which tends to have all the the abilities that both shaders can produce. Yes that would be the most logical answer. However I started using the dgs and dielectric for quite some time and I can still achieve certain looks that the mia_material struggles to produce. Anyways so I decided that since I couldn't seperate it at render time in the render passes then I would have to do it manually. However I came across another issue that caused pixels to go missing. That is where I came up with my topic where is Waldo, because for some strange reason the pixels decided to disapear once I applied a surface shader to everything that was not a dgs, or a dielectric shader. The solution to my horrible 1 hour dilema came to me while playing gears of war. I was pondering while shooting some bad guys when it hit me 'what about a contribuation map" and yes this fixed all my problems. I ahve to start thinking differently from now on because frmae buffers are cappable of doing some amazing things that where not possible before with out custom tools.

Or you can also go by what http://blue-riviera.blogspot.com

The solution is simple (but needs re-compilation of a shader, some artists might want to consult their fellow TDs :)). Modify your illuminance loop to look like the following:

while (mi_sample_light(&color, &dir, &dot_nl, state, *light, &samples)) {

if (dot_nl>0.0) // (1)
{
sum.r += dot_nl * diff->r * color.r;
sum.g += dot_nl * diff->g * color.g;
sum.b += dot_nl * diff->b * color.b;
// (...or your own illum stuff, whatever...)
}
}

(Sorry for the stupid non-formatting). The line marked with (1) first seems only as an obvious quick optimization, but it also filters out any incoming NaN values. (And these values do come in sometimes.)

If you add this dot_nl checking to your shader code, it will be (a bit) faster, and as a bonus it will get rid of those NaN pixels, too! :)

mi_car_paint_phen_x_passes

2009-01-11T23:21:00.000-08:00

So yesterday I finally figured out a slick way to separate the flakes from the mi_car_paint_phen_x mental ray shader. Even though frame buffers have been implemented in Maya 2009 the mi_car_paint_phen_x_passes which separates multiple render passes at render time can only go so far. For instance I noticed that the flakes and the specularity in shader can only be render together. If we go back to previous Maya, the car paint shader has almost always had problems with the flake scale if final gather or GI is turned on. When the flake scale is from .100 to .010 this causes tons of artifacts which cannot just be turned off by turning on jitter. This has been a huge frustration so for the past year or so I have turned off the flake scale down to 0. So then this brings up another question why use the mi_car_paint_phenomenon if the mi_car_paint doesn't have the flakes included isn't that a waste of time? no I have more control with mi_car_paint_phen I rarely even use the mi_car_paint. In Maya 2009 if the flakes are turned off to 0, then the car paint will render pure black as if it does not recognize the base color. Since I had no other choice but to use the mi_car_paint_phen_ with flakes turned on I figure it would be nice to give the compositor a pass where he could control the flakes, base color and lit color by rgb channels. This would means he could change the color, intensity etc of the flakes without having to re- render and the specularity would be separated from the flakes. Anyways here is my final result and a screen shot form Maya mi_car_paint_phen_ shader enjoy and I hope this might help. Remember this is done all in the shader not in the lighting.
*Note even though this helped me out there always tons of different ways to doing something.

-Finish 6+
-Paint scratches in photoshop
-Contribution pass for headlights, hood, and dgs materials
-(Done)Car paint shader flakes seperate by an RGB pass and pass contribution maps
-(DONE)Create displacement for the ground plane
-(DONE)Figure out how to get the old Mia materials to work in 2009 render passes
-(DONE)Set up new lights in the scene (hot spot)
-(DONE)Fix the shader for driver, rockets, and boosters

Framebuffers in Maya 2009

2009-01-09T18:57:00.000-08:00

I finally got frame buffers passes in Maya 2009 for Giant Freaking Robot. Converting each shader was not as complicated as I though it was going to be. However it was a tedious long job to get every thing to working correctly. Here are most of the buffer passes which will be combined with some render layers I have created manually. I expect for the compositer to let me know what he also might want from me.

* I am also working on a neat trick I though off to seperate the flakes from the car paint shader without damaging the specularity and color. Maya tend to crash but I think I might have it by tomorrow latest sunday. Well back to work check out Neon rockets they are sweet as peanut butter.
http://www.neonshaun.com/

Task list for Sunday

-Car paint shader flakes seperate by an RGB pass and pass contribuation maps
-Finish 6+
-Paint scratches in photoshop
-(DONE)Create displacement for the ground plane
-(DONE)Figure out how to get the old Mia materials to work in 2009 render passes
-(DONE)Set up new lights in the scene (hot spot)
-(DONE)Fix the shader for driver, rockets, and boosters

What are hot spots "dont get burned"

2009-01-08T02:09:00.000-08:00

In transformers behind the scenes Michael Bay mentions what are "hot spots". He explains that most CG cars don't look quite right because they are missing a hot spot. In my case we can call it a glare from the sun. Usually in the real world we tend to get hot spots either in exterior day light scenarios or even in a studio set up. This got me thinking and of course soon I added a hot spot to the car with the help of the mib_cie_d light from Mental ray. I also wanted to showcase the missiles, boosters, and gun which have been textured and shaded.

* Renderfarm at SCAD has implemented Maya 2009.

-Redo 13+ render layers
-Create displacement for the ground plane
-Figure out how to get the old Mia materials to work in 2009 render passes
-(DONE)Set up new lights in the scene (hot spot)
-(DONE)Fix the shader for driver, rockets, and boosters

There is always obstacles to tackle

2009-01-08T01:26:00.000-08:00

So this week we noticed that the animator had parent the rockets to the robot. So this means that when the robot would move the rockets would follow in mid air. This created all sorts of problems for Neon hence he was working on the rocket trails. Bum Joon our animator, rigger and modeler when back and fixed the problems. However I had to redo all the render layers because since we are working with multiple reference objects and about 600,000+ poly count Maya tends to be a bit buggy. On the other hand I hear SCAD might implement Maya 2009 into the farm. If this falls true I can do frame buffers and multiple outputs with the Mia material x and other mental ray shaders. Neon and I also discovered another problem that the resolution for the 2d camera move in Shake was not going to work, because the height did match up with the settings we had in Maya. However did problem has been resolved we ended up with a resolution of Width 1748 and Height 1560 this will compensate for the 2d camera move.

Here is my to do list

-Redo 17+ render layers
-Set up new lights in the scene (hot spot)
Will explain hot spot
-Create displacement for the ground plane
-Fix the shader for driver, rockets, and boosters

Giant Freaking Robot

2009-01-08T01:12:00.001-08:00

Neon and I will be fixing up GFR from Studio 1. However the deadline is on class 5, so we must work extremely fast and accurate. We have 2 weeks and half to get tons of stuff to work out correctly. Above is the where we left off in Studio 1 with Dave Kaul.

wwww.GiantFreakingRobot.com