Marmoset 2 Material studies

Hey guys, was doing some quick materials studies on an umbrella to test out some of the features of marmoset 2.  Mostly if they have the same limitations with depth of field and translucency. Unfortunately they do, but outside of that the program is very very strong and much better to use than before. Here are some screens:

My thoughts on first glance at Substance

Hey Guys,

So a few days ago I met up with the guys from Allegorithmic on my to look at adopting some of their pipeline for my own art. I ended up getting a complete walk through of Substance Designer 4 and the Substance Painter Alpha version by some of their tech guys and figured I would share some of my own thoughts.

The substance designer is a neat piece of kit. The idea is you make core materials (from cool noise generators or textures), then plug in a mesh, and a normal map (and maybe extras like a height map or world space normals, ao, cavity, etc) and use a node based system to get your final textures (it can figure out spec, gloss, diffuse all together). The neat part is that you can swap meshes and textures and get the same quality and material definition on a bunch of different assets using similar techniques (like making 150 guns really quickly). Non destructive workflow seemed to be pushed hard as well. If your AD needs one of the metals recalculated to have higher spec, you can batch export the materials all at once. It is also ready for version control software like perforce. The whole thing is built ready for physically based rendering, and supports custom shaders  (glsl) and channels.

The painter (which flaunts the new and fancy particle painting system seen here http://www.youtube.com/watch?v=wmEawPCAtEQ ) was also very cool. On the surface it features all the basics you would expect from a painting program, but with 3d painting on the forefront. You can paint layers, etc etc, and even paint in different channels based on materials (like painting gold material on a part of a model which automatically paints in the right gloss, spec, diffuse, etc). You can set up custom channels and pull materials right out of substance. One really neat thing is that nothing is ever rasterized. If your AD needs to change a material that you painted before layering up damage, overlays, etc you can update your model and not lose any work.  Finally the particle painting was neat, especially for explicit textures. Setup and adjustments of your own seemed very similar to basic scripting that sets up custom parameters for the artist to tweak.

Anyway, it was a cool meeting and I plan to check out the personal version myself. Here are some final thoughts/TLDR

-Targeted at AAA pipeline integration, used already in quite a few places
-Substance makes making lots of the same final textures easier by automating the material masking
-Priority of benefits is Productivity>Overall texture size(not run time) > actual runtime benefits (there are not many)
-the painter is looking to be a strong 3d painting contender
-particle painting is useful for explicit details

Some final thoughts:


For me personally, I think that if I were to tackle a large project or environment on my own (with some ready defined materials, it would be great. The painter I cannot wait for but I have yet to work with Mari much so I cannot say how it competes. 
It will however stay in the back of my mind for later :)

Ways to make a tree

This weekend I spent my time trying to settle on the right approach to making the trunk. My options in my head, and conclusions were this (keep in mind this is for a base mesh, not something to bake a normal from):

Speedtree -> don't own, cant export from demo version, UDK only viable for now, that one is out

Hand model -> viable, but slow, unwrapped can suck

Zbrush -> doable, but fine branch work gets tricky and slow

Paint effects -> control is limited, but has good UV options

I spent my weekend trying out each option and decided to get a test mesh in from both zbrush, and from paint effects. The zbrush one turned out alright, but was not worth it's time at all for the final shape (something I could have hand modeled in no time flat).

Something of interest to note was that decimating a sculpt to put my texture on, did let me fiddle around with what differences geo levels made. I tried 2 meshes, one at 1k tris and another at 4k and noticed that because I was rendering with tessellation, the levels made no difference. For my paint effects attempt, I kept things super low (especially on the smaller branches) to see what kind results super low poly work would bring in.

Again, it was all tests. For the most part I wanted to learn about different workflows and experiment with finding something that worked. My final call will be to use paint effects. Honestly the main reason for this is converting paint effects to polygons in maya makes for a UV perfect tree (a HUGE time saver). Looking at what I had at the end, there were some things to take note of:

-4 divisions is just too small for medium branches
-My texture is effective at certain scales for covering up mesh intersection (a big worry of mine)
-Variety in model silhouette can hide lots of repeating texture-ness (a real world I swear)
-Tessellation breaks down a bit with a global scale on tiny branches. I may make a non tessellated material for those branches.
-UV scale can change a bit between branch sizes but the small branch sizes need a smaller UV scale in general to get rid of some of that ugly noise.

Anyway tomorrow night Ill make my final base mesh pretty quickly (probably keeping it as a paint effect so I can add leaves through that method later) and move forward from there (I'm thinking twigs and leaves coming soon).

Finished Bark Texture

Hey guys, wanted to show the finished up bark texture. Tonight I took the information I learned from my lookdev before and used it to really polish up my final texture. First, I made the texture tiling, giving special attention to the final tile rate, as well as logical break up of space that doesn't draw the eye one way or another.

From my first test, I knew one of the biggest things I needed to do, was adjust the values and browns to make the bark feel more natural and saturated. I also knew what I wanted my final normal map to look like, and roughly how much information I needed in the height map. With final rough examples that was made fast, I was able to complete the entire final texture in far less time. A good lesson worth noting for the future. Test fast, learn and speed up your process with what you learned.

The final renders turned out great, and I may post some pics in the future with the Marmoset Toolbag 2. Overall I am pleased with the texture, and the micro detail holds up incredibly well (smart mix of normals and tessellation) Below are the final textures, and some beauty renders :) Tomorrow, comes making the tree itself!

Highrez Links Here:

 http://i.imgur.com/i8SsbTW.jpg

http://i.imgur.com/eCGmA6n.jpg

http://i.imgur.com/HMW8hEL.jpg

A return, and some fun pics

I am happy to say that after a few month haitus, I am officially back :)

Daily paintings will return (soon) as will more sculpting, and work on a few new projects. Tonight I spent my time doing some visual studies on a wood texture I was going to start soon.  The texture will (when done) be a tiling wood texture. Before I started though I wanted to get it in an engine (I chose marmoset for now) and see how different values held up and matched with lighting.

I chose 2 default images from marmoset (the forests) and one of my own (closer to my final image) to test out what I thought could be some of the final lighting. For the most part I was looking for darks that were too dark, and lights that were too light, as well as how the value range was getting shown once the image based lighting was hitting it.

The other thing I was doing was some testing with techniques to remove lighting from photos, as well as some fast normal creation techniques, and how they played with heightmaps. For a few hours play, the results were pretty good!

High rez link here: http://i.imgur.com/CD1GVdx.jpg