Original: http://www.onemanga.com/Air_Gear/264/18/

I think I like the cloth, that’s why I drew it out to study it. I intend to “study” more (do more copying type drawings) if I find the time, because I think drawing is a precious skill for an artist, I certainly want to be able to draw something nice out of my head, not using reference like what I just did.

GAHHH!!! (I suddenly realized) Did my pose looks too stiff? The original is more curve and dynamic~ I knew it… this makes me feel lame, noob, shame… I will watch out for that curvature next time!

A two man team developed this game,

…, I wouldn’t be surprised if it isn’t coming from Malaysians, but walao ehh!!!! This serious shit makes me so emo! Firstly, their arts are great, and the game engine looks great too. Then, they are using open source softwares completely. Then, as I read their blog, it’s getting clear to me that they are doing bunch of really cool stuffs. Like, helping the Ogre’s dev team to add PSSM shadow to their system (http://liquidrockgames.blogspot.com/2009/02/portals-shadows-and-materials.html)!!! I used to learn a little game programming years ago, and I tried to use Ogre, but not very successful, it’s hard for me to understand how to use it. These guys not just learn how to use it like a pro, they even read the source codes and write their own codes to contribute to the system (actually they is a single, lone, one, I repeat, one, one, one! programmer, not two, just one… sorry for lame jokes haha). I am really emo by that, simply impressive. They also have “Real time hologram distortion”, I don’t know what is that, but from the sound of it, it’s something cool. Using HDR in game, skybox, lightmaps, hovering physics, and so on… these are no kids stuffs.

Btw, when I see the name Lf3T-Hn4D, I immediately thought I knew this name somewhere, as I slowly remembered (with the help of Google), if I not mistaken, he is one of the founders of Game Developer Club of MMU. He’s the programmer for the game. Yap Chun Fei is the artist, from MMU too (http://www.blender-tutorial.com/interviste/aftershock/). Haha, really cool, kakilang (in Hokkien, the word here means we from the same place).

A very inspiring interview:
Interview with the Liquid Rock Games

Maybe I should print out some images of Aftershock and the interviews and paste it on my room wall… fight harder now, miica.

I love how the wireframe renders turns out. I learn it here: How to render Smooth wireframe in Maya, thanks tarjep :)

This my reference, grabbed from Flickr but I don’t remember the exact link: http://i282.photobucket.com/albums/kk243/miica37/2374265321_9ca99fb1a8.jpg

This is a work to be put into my demo reel to show my lighting and rendering skills. The texturing of it is very poor because I am running out of time… my family been pressuring me to find a job so I need to get things done. This image took very long to render (longest of all of my any previous renders), 10:31 on my PC, hmm… I think is because I am using the mia_material’s glossy but I am not sure. It took me about 1 day and a half, start working from just the 3d models. I also realized I still don’t understand the linear workflow, as I batch render the image and open it in Photoshop, it looks very different from what I see from Maya’s render view. So, in the end, I just save the image from Maya render view.

I use direct lighting, no FG or GI used. There are 3 area lights at the window side, and 4 spotlights, one from the top, one from the left side, one from the camera view side, and one on top of the two lamps. 7 lights in total. The preview render is rather quick because I halve the resolutions and also lower the raytrace sample to 2, previewing tooks 23 secs for this exact image.

I will continue to work on this image in the future when I find the time… I am going to create my demo reel now…

Another less pretty angle (render time is 3:10 at 512 x 288):

Update 18th November:

It’s a new render, tried hard to improve the overall looks. It still doesn’t look as good as I want, the worst is the render time, 15:30 mins, I have never waited this long before, and the AO is 12:30 mins, I wonder why…

I am working on this female body.

This is the settings for my sss fast skin shader:

I have also started to understand and use linear workflow. It just clicked to me yesterday… then I test a few scenes with simple shapes and one or two lights, the result is obvious that linear workflow is the better way to go.

Good readings:

* Linear workflow giving harsh falloff, what am I doing wrong?
* Tone and Gamma correction

For the render above I am using mia_exposure_simple (I dunno how to use the photographic one ah…). The settings:

How it looks when I remove the mia_exposure_simple:

The first one (with mia_exposure_simple on) looks like it’s been desaturated a lot, but look at the lighting! If I would continue working with the second one (with mia_exposure_simple taken out), I would try to add more lights to the scene and do lots of adjustment. For the first one, I just need to deal with the saturation by outputting the render to exr and tweak it in Photoshop (I didn’t do it for the first image btw, I just realized that it’s a bit desaturated, haha).

Post processed in Photoshop: