Unity3d game development game brought about by the problem

Yesterday there was a comment on an investor's decision to select the unity engine for the reason why the mobile team failed, but the tool itself was not guilty, but how to understand the tool, use the unity engine correctly, and you might need to understand the technical features of the engine's actual development process before choosing unity. and adapted to the game product type, enthusiastic reader fxcarl yesterday on this issue a special article, to help you understand unity game development, share experience, recommended reading.

Technical questions brought by code-driven

Fragmentation of the game. A very powerful feature of the U3D engine is the real-time compilation execution. This means that the current scene will go into an executable state regardless of when the execution icon is pressed. This has led to a game that often falls into a state of self-confidence in the development process. At the same time, it also led to the long-term fragmentation of the game content, and underestimated the difficulties that might be encountered in integrating game functions.

Resource management is a difficult point in the u3d engine. U3D resource management System because of cross-platform and operating system file system is decoupled, need to master Resources folder and Assetbundle Technical Talent Flexible Control game resource usage. But this work is often simply understood as placing resources under the game project folder, leaving the rest to the engine to take care of itself ...

Need to do their own data system. The vast majority of our domestic research and development work is data-intensive (strategy, business, card, KRPG), which is somewhat different from the game type Temple Run. Data-intensive games need to be designed and developed in a data-driven format, but the framework provided by U3D is a code-driven structure (which is very powerful for prototyping) and many times it can bog down the research and development team--it looks like the function has been developed ( Only to be able to work in the U3d object inspector, but not in the large-scale production phase (planning to hold the data table is not applicable to the game). The U3D engine itself does not provide any data support, and the data sheet either needs to be processed by itself or it needs to find its own embedded database solution.

The Network Connections section is actually similar. U3D itself integrates the network module is not for the large-scale C/s structure of the game design, often need to develop a set of client and server structure. Of course, you can also ask for middleware to solve ... But the easy puzzle is that u3d can use the. NET network mechanism to work like an end-trip, and also take a step back and be able to use the encrypted WWW mechanism when a simple page tour is processed. How to choose is a difficult problem, hasty too much perfection often in exchange for a distant future.

Testing the game developed by U3d is also a very troublesome process. The reason is that the scene/logic compositing editor, which is almost never crashing, can be executed at any time--it will allow the development team to count their current game completion and what tests they need.

The trouble with cutting-edge technology

Sophisticated dynamic lighting and complex material systems. U3d compared to other mobile platforms or web game development tools, often the most striking is its incomparable picture rendering effect. But behind the slick official demonstrations, it seems as if there are always unseen barriers that hinder the pace of other developers. In fact, the ability to harness u3d is beyond the ordinary imagination. U3d's rendering architecture is powerful enough to Unreal even CryEngine-level screen rendering quality is possible, but it does not package these systems but gives flexibility to developers. Does our program ape have control over the complexity of the rendering pipeline? Can our technical art guide our art to full play u3d ability? Does the art producer have the game content production that is competent for the so-called "next generation" accuracy requirement? Do these things belong to small teams?

Global illumination baking. This is a very useful u3d feature. All U3D teams should be flexible to use. But if you want to use it, then there is another difficulty-the art and scene production staff, and who is responsible for the more difficult. In addition, art must use very precise size to make objects in the scene, otherwise u3d will not be able to handle global Uvs correctly.

The chain of tools brings chaos

Various theories of the GUI system. Everybody's in the gutter. U3d's GUI system is too slow-the question is, is there evidence? On the one hand, a lot of people say that I did a lot of controls when I was testing, and it was really slow. On the other hand, you will find that the GUI system will bring some unnecessary rendering requests (Draw call). So everyone in the fight to do two things, one is to reduce the rendering request, one is to do everything possible to avoid the GUI. But in fact it's not that serious, whether selecting alternative middleware such as NGUI or the UI system that uses u3d directly does not have a huge impact--except for improper use, when the GUI becomes the focus of performance. Just NGUI or u3d the built-in UI has no very good UI tools-either too ape-oriented or too biased to the layout and inconvenient to add code functionality. It is necessary to develop some extension tools or workflows internally.

Problem with version number control. Asset Server or SVN is in fact a little bit of a situation that doesn't fit u3d. But the more important thing is to tidy up the internal structure of the file and often back it up. Proper use of U3d's command-line mode enables U3D project to proactively compile its own advertisement.

The ability to extend the functionality of the u3d itself. Due to the more complete function of u3d and neglect of the function of u3d itself is a common state, at any time to keep the person constantly optimize the function of u3d itself is very important, such as a variety of batch operations and so on. However, there is a premise that the extension tool needs to fully understand the tool, U3D is relatively powerful, so that very many members of the team will be afraid of learning, and the u3d as a few team members or special tools for the program ape-this is very problematic.

Requires a forward-looking inference capability

Each team that develops u3d games in China complains about U3d's Chinese font support issues and so on. But there are really few teams that actually use the U3D engine in a forward-looking perspective-in today's case, U3D 4.0 has been able to use dynamic fonts on whatever platform, supporting Unicode encoding-Chinese. Migrating from U3D 3.5 to 4.0 almost without making any changes to the project, and assuming that 4.0 would support dynamic fonts before, why not go to the official forum and focus on the development progress of each version number? Every game that is published at 2013 should not worry about font problems. ......

Keeping a focus on each version number u3d updates and future U3D features can reduce the problem of inventing wheels again, and keep a better attitude when encountering some difficulties. Direct mail developers can also be a great choice, so be sure to harass them more! It is generally possible to know what the future version number is likely to be updated from 3 months to 6 months in advance.

1 "Code-driven"

State Management

Assets Management

Data Management

Networking

Testing

2 Cuttingedge Techs

Dynamic Lighting & Complex materials (textures)

Lightmapping

Nav Mesh

Mecanim

DX11

3 Toolchain

Gui

Versioncontorl

4 Vision

Many other wonderful please click Http://www.gopedu.com/article

Problems with developing games developed by Unity3d Games