You should instead focus on the player experience which is the only result that matters. Both sound like improvements but often don’t produce any noticeable results. Another example would be artists crunching poly counts. I often see programmers optimizing various random bits of code, optimizing a UI foreach over some list from 10ms to 3ms. We can make optimization much more fun and satisfying with a simple plan. Also don’t keep switching between profiling targets, you need to decide on what your main target(s) are in order to make important decisions around performance goals. Don’t make random profiles in your editor, or simply on your own PC, as these don’t represent your target platform at all. It doesn’t make sense to go in blind and optimize code or art without having measured that these are your actual bottlenecks. Stop profiling or optimizing without a plan. No plan: Going in blind or making assumptions A game doesn’t have to run a solid 60 FPS during all phases of production, but at least keep it somewhat close to your target at all times to prevent a huge workload and overhauls in the last weeks. It’s really not doable to fix performance days or even weeks before your release, especially not because sometimes you might have to completely change how certain systems work. As usual there are exceptions to these rules but as long as you’re new to optimizing you’d do best to treat the following items as no-go. Let’s get started with some no-go’s to get some common mistakes out of the way. Therefore we focus solely on an analysis method you can use, this will inevitably point you to the right optimizations to make for your specific game. There is an overwhelming amount of optimizations you can make which are impossible to fully cover in this article. In this article I’ll present a plan that can be used as a base to profile any Unity game and I hope this can be a start to make your profiling work easier! As we’ve just wrapped up development of our latest game, optimized for 9 platforms, it was a good time to write down some of our techniques. Existing articles often provide a lot of pointers on what you could optimize, but don’t teach you how to figure out what makes sense for your game. It’s often done as a necessary rush job at the end of a production cycle, despite it being such an important subject. Few programmers build up this experience because there’s little time for optimization. It has always been hard to find programmers that have a good feeling for, and experience with, optimizing Unity games. My background and passion is the tech around game dev, anything from multiplayer to performance. For the last 13 years I’ve been running our own game studio M2H.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |