Odd Enough

This blog is not exclusive to 3D and game art. There are times like these where I will wax intellectual and talk about serious life stuffs.

One of the biggest reasons I love game development is that on any given day, the process of making something new quite frequently awards me with an opportunity to tax my mental capacity. This often leads to incredible amounts of frustration at first, but if I let myself ease into problem solving mode, it gets to be an extremely fun and gratifying experience. And when I do figure out what I have to do in order to work past it, I am filled with an almost overpowering sense of accomplishment and I wish I could just share my findings with the world. This feeling passes rather quickly as I am just as driven to push forward and find the next new challenge.

This process is not limited to game development or any of the work I do. This might sound strange, but I take this same approach with everything new. When faced with a challenge, I view it as a puzzle. A puzzle that needs to be analyzed and torn apart in my head. It’s like taking apart one the toys I had as a kid, find out why it stopped working after throwing it across the room, and reassembling it in order to improve my understanding of it. This rarely ever fixed the underlying problem of  the toy no longer working, but it did give me understanding. I assume most kids would have figured out that throwing a plastic electronic toy against a hard surface results in lack of function without all of that, but I preferred the scientific approach. It got to the point where I would take things apart even f they were not broken, just to appease my curiosity.

Recently I have been faced with a new challenge. What to do when you feel as if you’re staring down the face of your own future. This is a challenge that happens very rarely. In fact, I can only recall 4 times this has ever happened to me in my life. Yet something different happens when faced with such a unique situation: I am not haphazard about it. I am calm, not frustrated, and most of all, I feel confident that I already know the answer. I let my mind go through the motions, but clarity and lucidity often takes over.

When you make important life choices, there are no wrong answers or wrong decisions. There are simply paths that lead to one outcome or another. One path might be less ideal than the other, but making these decisions defines your life and ultimately, you. The best you can do is research every angle, every possibility, every little thing that might or could happen, and assume they won’t come to pass. I’m serious. Planning is the bane of your future. The best laid plans of mice and men oft go astray. Too fucking true. Choose a path that feels right to you and run with it, taking the curve balls that life throws at you right in the face. There is no ‘undo’ key, so you gotta keep moving forward. Each new bruise or scar just adds to who you are and your understanding, allowing you adapt faster and more easily to the next one. This is what I have learned since making the life choice to move away from friends and family and into an area where I had very few friends with me. My original plan for the move went haywire and ended up not coming to pass, but I stuck with my decision anyway. Turns out, I am REALLY glad I did!

Over this past weekend, I found myself explaining gloss maps and how they relate to UDK/game art to a few people on twitter. I find this completely hilarious because, up until a couple months ago, I had no clue how to use them either. Regardless, I’ll try to clear up their function and uses to the best of my knowledge.

So, if you’re reading this post, I’m going to assume you already have prior knowledge of specular maps. Would be freaking weird to start asking about gloss maps without knowing what spec maps are for. That’s just crazy. But if this is the case, and you are crazy, the easiest way to put it is that specular is reflection of dynamic, point, spot, and directional lights in relation to games. It doesn’t reflect the scene around the object, just the light (Update: I was informed that in some engines, the specular map also handles environment reflection). So when using specular maps in game art, the spec map dictates which areas on the model are going to be reflective of this light and by how much.

That’s simple enough, right? Of course it is. So we now have a way to control the specular reflection’s size and intensity. But hold on, not every type of material has the same specular values. We have surfaces like brick and fabric which are rough and have a very diffuse specular. And then we have smooth surfaces like metal, car paint, and crystal. How the hell do we control this madness?! Whoa whoa whoa… Calm yourself! Have a cup of tea, sit down and I will explain this for you.

*ahem*

Gloss Maps, that’s how.

Dramatization aside, gloss maps are grayscale representations of the specular widths of a material or grouping of materials. Find your texture map you wish to apply your gloss map to, and separate out how many different materials you have there. Say you have wood and glass in your texture, like on a door or window/picture frame. For correct specular representations (in the context of UDK) you could assign the frame and the glass two material IDs in your 3D suite of choice, and then make two different shaders to assign on your model in UDK, and tweak the specular power of each to match how it should look. It works, but it’s not ideal. To get around all of this extra work, you can use a gloss map to do that work for you. To demonstrate this, I’ve recruited one of the free textures from GameTextures.com to help in this example:

 

 

 

 

 

Shown above, you can see that the brighter areas on the gloss map are where the glass panes are. This produces a tighter specular width and gives a much more accurate appearance of specular than if you had no gloss map and a single specular width evenly applied to the specular areas. And since I bring up the subject of specular width, I feel inclined to explain that in more depth.

In explaining gloss maps to the people on twitter, it seemed that the concept really clicked with them when I mentioned specular width. Specular width is literally what it sounds like. In the case of most specular lighting models, they tend to be radial in nature, and the higher the gloss value (a scalar value of 0-100), the tighter or smaller the diameter of the specular is. I put together an image with one of my models to further illustrate this point:

 

 

 

 

 

 

 

 

 

 

 

 

As you can see, the higher the value, the tighter the spec width. Simple enough. If you’re confused as to where to plug the gloss map in within the UDK shader network, it’s the ‘Specular Power’ input as shown below:

 

 

 

 

 

 

 

 

 

 

 

 

One last thing I feel is a good practice to get into when working with specular and gloss maps is to bitpack them. If you aren’t aware of the term “Bitpack”, don’t panic! It’s very possible that I made that up, but it made sense to say that when I first figured it out, so I’d use that term tentatively. Bitpacking is a method used to pack multiple texture maps into one texture file by loading greyscale representations into R, G, B, and A channels. This is useful for saving texture space and for organizational purposes. Within UDK, you are able to split out each channel separately and control color and intensity at will. I tend to pack my emissive map into the Alpha of my Diffuse, and change color/intensity within the shader editor. In this case, it’s generally a good practice to pack the gloss map into the alpha channel of the specular map. It’s not ideal to do the reverse, mainly because gloss maps are purely grayscale while specular maps are generally based off of the diffuse and thus have RGB values.

To sum up: Specular = reflection, Gloss = Approximation of microfacet roughness

Alright, I need to get back to work, but I hope I was able to clear things up to a point. If you have any further questions about this or want to tell me how wrong I am, feel free to yell at me via @odd_enough on twitter.

Throwing up this quick post just to say that I’m postponing Week 3 until the week after the Global Game Jam. I figure there are probably more people than just myself doing that, so eh. In the meantime, you can continue to work on Week 2′s theme.

So Week 1 of my thematic modeling exercise was an unexpected success! There were even a couple contributors who I don’t even know personally. This makes me happy when random people see my idea and jump on it. I will post the Week 1 recap shortly. Waiting on a couple people to submit their results.

Going to keep this simple this time; No wall of text to sift through to find the pertinent information. Here goes:

 

• Subject – Forms of Communication
• Time Period – Any
• Creativity – Encouraged

 

Lots to choose from! I also have a lot of comedic ideas for this. What to do, what to do…

This recap was posted a little bit later than I was hoping, but I wanted to give a couple people to get something of what they made in, finished or not. Regardless, I am happy with the turnout for the first go of this exercise!

The art of those who participated (designated by twitter handle):

 

@odd_enough:

 

 

 

 

 

 

 

 

 

 

@ruthmadeline:

 

 

 

 

 

 

 

 

 

 

 

 

@KevinO_Designs:

 

 

 

 

 

 

 

 

 

 

 

@demafleez:

 

@chromashift:

 

 

 

 

 

 

 

 

 

@mystichobodev:

 

 

 

 

 

 

 

 

 

 

 

 

Thanks to everyone who contributed! Looking forward to seeing what people make for Week 2.

To put this as simply as I can, the goal of this exercise is to create a single model over a 1-week time span that fits with that week’s chosen theme. Doesn’t sound too bad, right?

Right. And that’s the point. I don’t want to make it so complex that you’ll feel overwhelmed before you’re halfway through reading this. That’s stupid. That said, there are some extra details involved that are designed to make it a bit more interesting and fun, and make sure there are some guidelines to keep things moving. Making it too restrictive isn’t fun, and making it with no restrictions at all is just asking for trouble. So here is the rundown:

 

Each week, starting and on Sundays and ending the following Sunday, a theme with a short descriptor is chosen and posted on this section of my blog (with a quick link to it up top on the nav bar), which will be linked on Twitter. The theme can be any variety of things such as a location, weapon types, an overarching style with subset parameters, and others.

As an example theme, I would say something like:

• Style – Sci-Fi
• Object – Weapon
• Object type – Rocket Launcher

You would then use the information available to you to make your own model. This forces you to use abstraction in your designs, as well as keeps things open enough for creative thinking. There is no restrictions on how you go about modeling your object; Just that it needs to be high poly, untextured at minimum. If you wish to take it through a full workflow of your choosing, that is entirely up to you, but not required. Keep in mind, that regardless of how far you want to take it, I will only accept submissions for a given theme at the end of its time span. No late entries. The scope of the model is up to you, so plan your model accordingly.

To be clear, this is not a contest! This is a way for you to challenge yourself personally. Expand your horizons past what you are normally accustomed to. If you want to use only ZBrush for this as an added extra challenge, you may; It’s your choice. I am simply providing a conduit for direction.

I wanted to keep this exercise to be very open. I didn’t want to hold a specific list of people (or even myself) accountable to finish EVERY theme. If you feel like you are able to participate for a given week, awesome! If not, no hard feelings. The theme will continue to be posted, regardless if anyone has time to do it that week.

For now, I have no collaborative system set up other than Twitter and this blog. So when you complete your model for a theme, make a comment with the image and/or turntable of your model in the corresponding post. When posting WIPs on Twitter, use the hashtag “#ThemeChallenge” followed by the week number. So this week would be: “#ThemeChallenge1″. I’m considering posting the results of each week on the Polycount forums.

If you have ideas for themes, pass them my way and I will consider them for future use.

 

And off we go! This weeks theme:

• Style – American Old West
• Descriptor - Items found on an Old West ranch

Yes, you might need to do some research, but I counted at least 10 unique objects to work off of just from a quick Google image search.

What are you still doing reading this? GO GO GO!

For the people who actually read my blog, you’ll know that I tried an exercise last year where I attempted to produce a 3D model for each day of the year. I only made it up to 22 models before a bad batch of raw milk landed me in the hospital. This set me back quite a bit and I eventually gave up. Regardless, a model per day is more difficult than it sounds if you already have a full-time job.

I was having a conversation with a fellow artist during Christmas vacation and it triggered an idea to try a similar experiment but with a slight twist. What if I open up the exercise to allow multiple artists to attempt to model the same thing, expand the deadline to a week, and record the results on a sub-section of this blog? Hmm. Worth a shot.

The initial response on twitter was better than I expected, so here I am starting to plan in the only way available to me for now. So let’s start this off with some recurring themes I’ve noticed in both my peers and myself:

 

- We often find ourselves second guessing ourselves not only in the quality of our work, but in what we’re currently working on. As a result, we start many personal/portfolio projects, but only get around to finishing a fraction of them. A lot of us have the attention spans of gerbils.

- Personal attachment to our artwork is a bit problematic. It’s not a bad thing, but it can get out of hand. Having experienced this both first and second hand, I can say that it does get in the way of making steady progress on a project. You need to be able to detach, look at it objectively, decide when you’ve worked on something long enough, if it’s acceptable, and move on.

- Completion time is another big one. Yes, I know we’re artists, and good art takes time, but we’re also employed or looking for employment. Regardless of your status, we need to incorporate prompt completion into our standard workflow.

 

One of my own personal goals for this year is to work on the issues listed above. I feel that I could become a more efficient and effective artist if I can start to chip away at those things at the very least. I doubt it would hurt anyone else to strive for similar results. How you go about doing that is up to you. But if you wish to participate with this, let me know on twitter, or make a comment on this post.

I will be posting information on the first exercise on Monday, January 7th, 2013.

 

One thing I always wonder about when looking at various artist’s portfolios is how long it takes them to complete an asset. I see really awesome art and they show texture sheets, high poly and low poly versions, wireframes, programs used, etc. But no time spent working on it. Do most potential employers not find this to be crucial to their decision-making? I understand why people don’t usually post it voluntarily. Most of the time, it takes people forever to make something look good. This trend of withholding that information needs to stop.

I’ve caught glimpses of certain artists sharing their average asset completion times, but most of them are established artists, or explaining details of their art department workflow for an already released game. Maybe this is not important for AAA game studios, but for indie studios, completion time is crucial. Indie budgets are less than flexible, and the notion that I keep hearing about that you can’t make a AAA PC title for under $10 mil (not including marketting) is 100% unacceptable.  While the dev time isn’t completely taken up by asset creation, it is a large factor. As artists, we should not only strive for quality, but we should also strive just as much for speed.

Games with dev budgets of multi-millions should NOT be something positive to brag about. Hearing studios comment on those numbers in the media like they were proud of it makes me want to gag. Our industry should be striving for efficiency. And people wonder why many big studios are falling on hard times.

From now on, I will be transparent about my creation/completion times. We need to raise our bars.

LVL26! NEW ASSETS! TEAM MEMBERS! LENS FLARES! CAPSLOCK!

So I just made a blog post earlier this week on the LVL26 Blog about my asset creation workflow. Had a lot of fun with it, but didn’t go into as much detail as I had hoped. Maybe that’s a good thing. It probably would have bored most non-artists to death. If I ever have the time (hah!), I’ll go more in depth here.

Speaking of not having time, I want to describe each of the assets I’m about to post in detail, but I can’t. So I’m just going to post images. Textured assets are in-game UDK low polys: Node (2193 tris), Barrier post (1668 tris), Small crate (108 tris), Big crate (544 tris)

 

I started working with two more modelers on LVL26: Amanda Lange and Gabriel Cabrera. I met Amanda at the 2012 Global Game Jam in Harrisburg, PA, and Gabriel is a very promising young artist trying to learn the ways of modeling.

Also, I did promise you lens flares, so here you go:

- J

Mostly finished the high poly of the turret in the previous post, was put aside for work on a game prototype.

Since then, I have been hard at work creating assets and materials. A couple things I can show off is my UDK master material tree and an example material to go with it:

So that’s that. Will try to update this more, but seeing as I keep forgetting… I’ll update it when I remember to