![]() CinematicsĪ slightly more demanding type of feature is cinematic-only one. They not only looked better, but also were much faster – obviously worked only because of those special conditions. Jungles were “cut off” from the rest of the open world by special streaming corridors and we completely replaced the whole lighting in them! Instead of relying on tree shadow maps and global lighting, we created fake “caustics” that looked much softer and played very nicely with our volumetric lighting / atmospherics system. Notice the caustics-like lightshafts that were key rendering feature in jungle levels – and allowed us to save a lot on the shadows rendering! Source: Assassin’s Creed 4: Black Flag promo art. In this case, a feature doesn’t need to be very “robust”, and often replaces many others.Īn example could be lighting in the jungle levels in Assassin’s Creed 4: Black Flag that I worked on. It can be some visual special effect happening in a single scene, game intro, or a single level that is different from the other ones. The next level “up” in the difficulty is creating some special features that are one-off. The rest? Hack away, write one-off code – just don’t have expectations that turning a tech demo into a production ready feature is simple or easy (it’s more like the 99% of work remaining). The main considerations will be around performance (a choppy tech demo can be seen as a tech failure) and working very closely with artists able to show it off. You can actually retrofit everything: from the demo concept, art itself, camera trajectory to show off the technology the best and avoid any problems. If your whole product is a demonstration of a given technique (whether for benchmarking, showing off some new research, artistic demoscene), most of the considerations go away. The categories of “use cases” deserve some explanation and description of “severity” of their constraints. I am going to describe first the use-cases – potential uses of the technology and how those impact potential requirements and constraints. Very important note – none of the “obstacles” I am going to describe are deal breakers.įar from it – most successful tech that became state of the art violates many of those constraints! It simply means that those are challenges that will need to be overcome in some way – manual workarounds, feature exclusivity, ignoring the problems, or applying them only in specific cases. Role of the fundamental research is inspiration and creating theory that can be later productionized by people who are experts in productionization.īut if you are a pure researcher and somehow got here, I’ll be happy if you’re interested in what kinds of problems might be on the long way from idea or paper to a product (and why most new genuinely good research will never find its place in products). Note that I didn’t place “pure” academic researchers in the above list – as I don’t think that pure research should be considering too many obstacles. Some concepts might be too technical and too much jargon, but then feel free to skip those. People who are excited and care about game graphics (or real time graphics in general) and would like to understand a bit “how sausages are made”.Hardware engineers and architects working on anything GPU or graphics related (and curious what makes it complicated to use new HW features),.Technical directors and decision makers without background in graphics,.Tech artists and art leads / producers,.Rendering engineers, especially ones earlier in their career – who haven’t built their intuition yet,. ![]() ![]() Students in computer graphics and applied researchers,.I tend to think of this topic as well when I hear discussions about how “photogrammetry, raytracing, neural rendering, will be a universal answer to rendering and replace everything!”. □ And yes, I will come back to vegetation rendering a few times in this post. The post is also inspired by my joke tweet from a while ago about appearing smart and mature as a graphics programmer by “dismissing” most of the rendering techniques – that they are not going to work on foliage. I have seen statements like “why is this brilliant research technique X not used in production?” both from gamers, but also from my colleagues with academic background. Many of those challenges are easily ignored – they are real problems in production, but not necessarily there only if you only read about those techniques, or if you work on pure research, writing papers, or create tech demos. I will base this on my personal experiences, working on Witcher 2, Assassin’s Creed 4: Black Flag, Far Cry 4, and God of War. This post will cover challenges and aspects of production to consider when creating new rendering / graphics techniques and algorithms – especially in the context of applied research for real time rendering. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |