I’ve been growing more and more curious over the past couple weeks regarding the new flashy graphics features and implementations that reside in Unity 5.
Do you feel too unfamiliar with them? Do they, well, frighten you in a way that keeps you from trying them out or studying them? If an Asset Store publisher is using the features, is that something that discourages you from purchasing?
If someone were able to explain concepts like PBR, Screen Space Reflections and Ambient Occusion, Color Grading, Dynamic and Static lighting scene setups, even Normal Mapping, in a clear and simple way, do you think you’d use them more?
I’m considering doing just that with a series of solid, professional videos about these features, but with a sharp focus on explaining them in very simple, accessible, artist-centric terms without boring the hell out of you. What do you think?
Remember, it’s much less about HOW they work, it’s about WHY you should use them, what things you need to know, and how it’s actually easier than you think… Sort of a ELI5 series for Unity.
Yes, I’d love some introductory/overview material on those. If I found a well-written article (or series of articles) with useful illustrations, I’d totally read that.
But I doubt I’ll watch the videos… I’m just too busy. I don’t have time to spend 20 minutes watching content I could read in 5.
Agreed, that’s why they wouldn’t be long. A companion document would probably be useful as well. From an end-user perspective, I hate having to absorb a load of information that I deem useless to what I’m trying to learn.
Google is your friend here, there’s some short and lengthy explanations of all of these at all different levels of depth out there. And some of the techniques require a lot of understanding of several parts of render to make sense of beyond the basic explanations.
However here are a few.
PBR - Physically Based Rendering (sometimes Physically Based Shading)
This is a blanket name for several different techniques to try to better replicate the way light actually reflects off of surfaces, both for diffuse illumination and specular highlights, though often more focused on specular highlights. Has the benefit of generally requiring less work for “difficult” surfaces and looking good under a wide range of lighting scenarios.
SSR - Screen Space Reflections
A technique that lets surfaces reflect stuff that are on screen, hence “screen space”. If it’s not in view before it does reflections, it won’t appear in the reflections. Generally looks much closer to a “real” reflection, or a reflection you could get using a non-realtime renderer, which are difficult or impossible to replicate consistently with other real time techniques and aren’t limited to flat surfaces.
AO - Ambient Occlusion
Approximating the blurry shadows from ambient or very diffuse lighting like from a perfectly overcast day. A common comment about AO is it helps “ground” objects; things look like they’re close together / touching especially in areas without direct lighting and shadows.
Color Grading
Any kind of adjustment to the colors of the scene after rendering. Think Instagram filters. Do you want your game to look like it takes place in the 60s (desaturated colors, slightly reddish), or the 1890s (black & white or sepia), or just want everything a little more blue?
Dynamic & Static Lighting
Dynamic lighting can move the light source, and objects that move around can cast shadows. Static lighting the light source can’t move and objects that do move won’t cast shadows and won’t receive shadows cast from objects that don’t move. Mixed lighting the light source can’t move, but dynamic objects can cast shadow but still can’t receive shadows from objects that don’t. Static lighting is usually the fastest to render but requires a baking time before it can be used. Dynamic lighting is usually slower to render or maybe has to not have shadows and nice ambient lighting, but can work with stuff that changes and moves without needing baking making it easier to work with or use with procedural content.
Normal Mapping
Modifying the direction a surface appears to be oriented using a texture instead of polygons. Allows you to use lower polygon models while still getting the appearance of higher polygon counts. Can sometimes be reused to make the same geometry look completely different, just like taking a flat white wall and adding a brick or wood diffuse texture, also using a brick or wood normal map texture allows it to look like the geometry has been changed too.
Seriously though, the idea with Unity has to be “discover what you need, when you need it” because it’s actually impossible for staff to know every aspect, let alone the end user. Engines are absolute beasts in today’s age. It’s normal to be overwhelmed, if you think you should know it all but use very little of it. Chances are, you don’t need a lot of it.
Things like combined cinematic post take away the need to know at face value, then if you’re having a problem, dive deeper. That’s my advice for people in general here.
Yeah, but it’s important to know what you don’t know.
For example:
This sounds like exactly what I need to solve some problems on one current project, and I had no idea such a thing even existed. I was starting to ponder how to do it myself with shader tricks etc., because I didn’t even know there was this whole AO thing that I didn’t know about.
Now that I know about my AO ignorance, I can go do something about it!
Heheh. While I appreciate the long explanations and research you’ve done for me, that’s not the reason for the post. I was wondering if people would be interested in a series that goes into this detail with examples. I already know these particular topics. Thanks though!
I think it would be helpful for the community at large. Though I think it’s important to show benefits and drawbacks from using ‘extra’ features of the engine, how best to use them, why one might consider using them, why one might consider looking for alternative approaches / why one might not want to use a technique/feature, how to set them up for ‘general’ use, and pitfalls to watch out for.
For me getting started with a new topic often works best watching a video which mostly can be sped up 1.5 or even 2 times the normal speed and still get some meaning out of it. Some subjects are just quicker to learn from a video than a sea of screenshots.
Searching for a quality video on a very specific subject can be a task in itself though so then a written tutorial is the best option.
