Unity 3.4 is nearing release, and we wanted to share some the features that will be included and also share with you a roadmap for what we are working on this year.
Unity 3.4即将发布,我们希望分享一些将包含的功能,并与您分享我们今年正在努力的路线图。
We’re about to enter release candidate 1 with Unity 3.4, which is a feature + bugfix release. Here are some highlights:
我们将使用Unity 3.4输入候选版本1,这是一个功能+错误修正版本。 以下是一些要点:
Procedural textures are built directly into Unity. Substance procedural textures can be tweaked in Unity. You can even update your textures at run time. It can be used for anything from ageing effects on textures to customizable characters.
程序纹理直接内置到Unity中。 物质程序纹理可以在Unity中进行调整。 您甚至可以在运行时更新纹理。 它可以用于从纹理的老化效果到可自定义字符的任何事物。
Directional lights got a new shadow projection mode called “Stable Fit”. Now there’s no more shadow boundary shimmering when rotating the camera! Regarding optimizations, we cull shadow casters much better now, which means fewer draw calls. Shadow culling can also use occlusion culling data.
定向光获得了一种新的阴影投影模式,称为“稳定拟合”。 现在,旋转相机时不再有阴影边界闪烁! 关于优化,我们现在可以更好地剔除阴影投射器,这意味着更少的绘制调用。 阴影剔除也可以使用遮挡剔除数据。
Skinning performance is 2-3x faster on PCs due to SSE2 optimizations and multicore improvements. There’s also multithreaded skinning for the latest iOS and Android devices.
由于进行了SSE2优化和多核改进,因此剥皮性能在PC上快了2-3倍。 最新的iOS和Android设备还具有多线程外观。
This is a cross-platform feature including iOS and Android that’s perfect for downloadable content. Asset bundles and even complete scenes can be downloaded on demand and cached locally. The caching API gives you full control over when to download a newer version of the asset. Using the caching API dramatically reduces memory usage and has a huge impact on load times.
这是包括iOS和Android在内的跨平台功能,非常适合可下载的内容。 资产包甚至整个场景都可以按需下载并本地缓存。 缓存API可让您完全控制何时下载资产的较新版本。 使用缓存API可以大大减少内存使用量,并且对加载时间有很大的影响。
Unity now gives you eplicit control over the execution order in your scripts. Awake, OnEnable and Update calls are sorted by execution order. Execution order can be defined on a ‘per script’ basis in an easy-to-understand dialog.
现在,Unity使您可以轻松控制脚本中的执行顺序。 唤醒,OnEnable和Update调用按执行顺序排序。 可以在一个易于理解的对话框中基于“每个脚本”定义执行顺序。
Pre-computed bounding volumes for skinned meshes & animations let you have more animated characters in your scene.
蒙皮网格物体和动画的预先计算的边界体积使您在场景中拥有更多的动画角色。
We implemented various graphics optimizations, ranging from faster deferred lighting (due to more compact light shapes & better occlusion culling) to faster OpenGL ES 2.0 and more mobile optimized shaders. Last but not least: Terrain works on iOS and Android now.
我们实施了各种图形优化,从更快的延迟照明(由于更紧凑的灯光形状和更好的遮挡剔除)到更快的OpenGL ES 2.0和更多移动优化的着色器。 最后但并非最不重要的一点:Terrain现在可在iOS和Android上运行。
There’s a big upgrade to image effects & water. Both have been optimized for performance and, at the same time, visual quality has been improved. They’re also easier to tweak for artists.
图像效果和水有了很大的提升。 两者都针对性能进行了优化,同时,视觉质量也得到了改善。 它们也更容易针对艺术家进行调整。
All console versions are now released at same time as 3.4. and projects can easily be moved between all platforms.
现在,所有控制台版本都与3.4同时发布。 并且可以在所有平台之间轻松移动项目。
Rochard, our first PS3 game, has just gone through the Sony submission process.
我们的第一款PS3游戏Rochard刚刚完成了Sony提交流程。
You no longer need to code to add icons to your game objects. Now you can simply set the icon in the game object inspector to quickly see and select things in your scene view. If you want to see icons for tools like pathfinding nodes, just set the icon on the node script in your project and they’ll all show up in the scene.
您不再需要编写代码即可向游戏对象添加图标。 现在,您只需在游戏对象检查器中设置图标即可快速查看并选择场景视图中的内容。 如果要查看诸如寻路节点之类的工具的图标,只需在项目中的节点脚本上设置图标,它们就会全部显示在场景中。
You can also toggle the display of handles on a per-component basis so now there’s no need to hold back when writing custom editors.
您还可以在每个组件的基础上切换控柄的显示,因此现在在编写自定义编辑器时无需停下来。
If you select a collider and hold down Shift, small dots for resizing the colliders will appear. This makes it really quick to set up efficient collision levels so they run really fast on mobile devices.
如果选择对撞机并按住Shift键,将出现用于调整对撞机大小的小点。 这使得设置有效碰撞级别的速度非常快,因此它们在移动设备上的运行速度非常快。
Unity 3.4 also has loads of smaller improvements, shine-ups and fixes.
Unity 3.4还包含许多较小的改进,改进和修复。
This year we’re focusing on enabling you to push all platforms to their limits. Unity is the most optimized game engine for mobile devices and we’re working hard to also make this true of all other platforms. It’s a big step for Unity to support AAA productions. We have tons of performance optimizations, better multi-threading and a big graphics engine upgrade. At the same time, we are also improving work flows, especially when working in larger teams. Unity 3.5 is a late summer / fall 2011 release.
今年,我们专注于使您将所有平台推向极限。 Unity是针对移动设备最优化的游戏引擎,我们正在努力使所有其他平台都实现这一目标。 对于Unity支持AAA生产而言,这是迈出的一大步。 我们进行了大量的性能优化,更好的多线程和大型图形引擎升级。 同时,我们也在改善工作流程,尤其是在较大的团队中工作时。 Unity 3.5是2011年夏末/秋季发布。
Please note that the features in this roadmap are not commitments! They may or may not end up in a final release. If a feature isn’t ready in time it will not be included in the release.
请注意,本路线图中的功能并非承诺! 他们可能会也可能不会最终发行。 如果某个功能未及时准备就绪,它将不会包含在版本中。
We will be moving most of the rendering workload and driver overhead away from the main Unity thread and onto another CPU core. This is an ongoing project that is currently working on the PC but we plan to make it available on all multicore platforms, from desktop to mobiles. Plus, we’re going for wider multicore utilization (using as many CPUs as there are) for tasks like culling and particle systems. How much of this is ready for 3.5 and on which platforms is still an open question, but we can say that the performance gains look very promising!
我们将把大部分渲染工作量和驱动程序开销从Unity主线程转移到另一个CPU内核。 这是一个正在进行中的项目,目前正在PC上运行,但我们计划使其在从台式机到移动设备的所有多核平台上可用。 另外,我们将为诸如剔除和粒子系统之类的任务提供更广泛的多核利用率(使用尽可能多的CPU)。 其中有多少准备好用于3.5,以及在哪些平台上仍然是一个悬而未决的问题,但是我们可以说,性能提升看起来非常有希望!
Unity will automatically generate navigation meshes from your level geometry. Beautiful, natural-looking crowd simulation using RVO and PLE algorithms wrapped in a simple API. Agents can find paths to target locations with built-in crowd simulation, or can be moved directly on the NavMesh in a similar way to the character controller.
Unity将根据您的关卡几何自动生成导航网格物体。 使用包装在简单API中的RVO和PLE算法进行美观,自然的人群模拟。 座席可以使用内置的人群模拟功能找到到达目标位置的路径,或者可以以与角色控制器类似的方式直接在NavMesh上移动。
We re-wrote our particle system from scratch. The new particle system is completely curve and gradient driven, and is composed of modular blocks to boost functionality. Also, it utilizes multicore CPUs on all platforms!
我们从头开始重新编写了粒子系统。 新的粒子系统完全由曲线和梯度驱动,并由模块化模块组成以增强功能。 此外,它在所有平台上都使用多核CPU!
We’re rewriting the in-game GUI system from the ground. We’ve made it simple and intuitive to create modern game interfaces with tons of animations and other effects. Everything is assembled in a visual GUI editor. A lot of effort is going into optimizing the GUI system for minimizing run-time overhead through automatic texture atlases and aggressive coverage-based batching.
我们正在从头重写游戏中的GUI系统。 我们使创建带有大量动画和其他效果的现代游戏界面变得简单直观。 一切都在可视化GUI编辑器中进行组装。 通过自动纹理图集和积极的基于覆盖的批处理,为优化GUI系统以最大程度地减少运行时开销,人们将进行大量工作。
Light probes are an advanced technique for lighting dynamic objects and characters with high-quality baked lighting. We capture incoming direct and bounced lighting within a scene at a number of points (light probes). As a character moves through the scene, nearby probes are picked, interpolated, and the result is then passed to the shader to light the surface. It’s a hard problem to do all this efficiently but thanks to our tetrahedra-based space division, the technique is extremely fast even on mobiles and has a very low memory footprint. For more details check out the video and blog post.
光线探测器是一种先进的技术,可以通过高质量的烘烤照明来照亮动态物体和角色。 我们在场景中的多个点(光探测器)上捕获传入的直接和反射光。 当角色在场景中移动时,将拾取附近的探针并进行插值,然后将结果传递到着色器以照亮表面。 有效地完成所有任务是一个难题,但是由于我们基于四面体的空间划分,该技术即使在移动设备上也非常快,并且内存占用量非常低。 有关更多详细信息,请查看视频和博客文章 。
To add more flexibility to our lightmapping solution we’re adding the option to bake directional lightmaps (radiosity normal maps and similar schemes). That will also enable bump and specular effects on surfaces lit by indirect light only.
为了增加我们的光照贴图解决方案的灵活性,我们添加了烘烤定向光照贴图(辐射度法线贴图和类似方案)的选项。 这还将在仅由间接光照明的表面上启用凹凸效果和镜面效果。
Built-in Level Of Detail support. LODs are currently authored manually in your favorite modelling tool but we’re also experimenting with automatic LOD generation techniques.
内置的详细级别支持。 目前,LOD是在您最喜欢的建模工具中手动编写的,但我们也在尝试使用自动LOD生成技术。
Intuitive editor tools for setting and inspecting LODs are included as well as integration into the pipeline to automatically create LODs based on naming conventions. We’re also offering support for fading level of detail transitions using screen-space dithering. Get more information in this blog post.
包括用于设置和检查LOD的直观编辑器工具,以及集成到管道中以基于命名约定自动创建LOD的工具。 我们还为使用屏幕空间抖动的细节过渡的淡入度级别提供支持。 在此博客文章中获取更多信息。
We’re working on fully integrated version control support for both Perforce and Subversion. There’s a complete UI including support for file locking.
我们正在为Perforce和Subversion提供完全集成的版本控制支持。 有一个完整的UI,包括对文件锁定的支持。
Unity will now write all data in a text-based file format for scenes, prefabs, materials and other binary files in your project folder. The format is based on YAML and is optimized for being easy to merge. Multiple team members can work on a scene at the same time and merge the resulting changes afterwards.
现在,Unity将以基于文本的文件格式将所有数据写入场景,预制件,材料和其他二进制文件到您的项目文件夹中。 该格式基于YAML并经过优化,易于合并。 多个团队成员可以同时在一个场景上工作,然后合并结果更改。
The prefab system has been rewritten. It allows you to add and remove components without breaking prefab connections. Materials can be instantiated with the prefab instance. This paves the way for nested prefabs.
预制系统已被重写。 它允许您添加和删除组件而不会破坏预制连接。 可以使用预制实例实例化材料。 这为嵌套式预制件铺平了道路。
A new allocator framework helps us reduce memory usage and fragmentation and improve performance. Now there’s a much better memory overview in Unity’s Profiler.
新的分配器框架可帮助我们减少内存使用和碎片并提高性能。 现在,Unity的探查器中有一个更好的内存概述。
We are working on a pluggable social API with backends for Facebook and Apple Game Center. Companies like OpenFeint or publishers with custom social platforms can easily create their own implementation. This makes it easier to port games between platforms.
我们正在开发具有Facebook和Apple Game Center后端的可插入社交API。 像OpenFeint这样的公司或具有自定义社交平台的发布商可以轻松创建自己的实施。 这使得在平台之间移植游戏变得更加容易。
We are working on low latency audio buffer access. This will enable you to write your own filters, effects or make the audio data precisely affect your application or let your application dynamically mix and procedurally generate audio samples.
我们正在致力于低延迟音频缓冲区访问。 这将使您能够编写自己的滤镜,效果或使音频数据精确地影响您的应用程序,或者让您的应用程序动态混合并以程序方式生成音频样本。
Microphone support will also be introduced across all platforms. Microphone input can be accessed transparently as an AudioClip.
还将在所有平台上引入麦克风支持。 麦克风输入可以作为AudioClip透明访问。
We’re adding support for streaming video from cameras or other input devices into a texture to allow you to develop augmented reality games, or for uploading pictures for integration with social networks or other online services. This was implemented at our last NinjaCamp week.
我们添加了对将来自摄像机或其他输入设备的视频流化为纹理的支持,以使您能够开发增强现实游戏,或上传图片以与社交网络或其他在线服务集成。 这是在我们上一个NinjaCamp周执行的 。
Umbra’s sPVS occlusion culling system has been rewritten from scratch. The algorithm now guarantees conservative PVS results and baking PVS is around 10x faster.
Umbra的sPVS遮挡剔除系统已从头开始重写。 该算法现在可以保证保守的PVS结果,烘烤PVS的速度快10倍左右。
As usual, we fully expect Unity 3.5 to deliver tons of other small features, improvements and fixes.
与往常一样,我们完全希望Unity 3.5能够提供大量其他小功能,改进和修复。
Again, the fact that we are working on it is not a promise that it will ship in a matter of weeks. We just want you to know what we’re trying to do. Some plans might change!
同样,我们正在努力研究这一事实并不意味着它会在几周内发货。 我们只想让您知道我们要做什么。 一些计划可能会改变!
As announced, we are hard at work to add support for building a Unity project that can be played back by a browser equipped with the Flash “Molehill” plugin. We’ll bring as many Unity features to the Flash Player as we can, and aim to be the very best tool available for people who want to target the 3D functionality in the new Flash Player.
正如宣布的那样 ,我们正在努力添加对构建Unity项目的支持,该项目可以通过配备Flash“ Molehill”插件的浏览器进行播放。 我们将为Flash Player带来尽可能多的Unity功能,并力争成为希望在新Flash Player中定位3D功能的人们的最佳工具。
Everyone loves Unity’s completely automatic asset pipeline and the ability to quickly modify any asset in your project folder without jumping through hoops.
每个人都喜欢Unity的全自动资产流水线,并且能够快速修改项目文件夹中的任何资产,而不会遇到麻烦。
But when working in larger teams, everyone has to import the same assets again and again. If it could only be 3000x faster to import all assets… We have a solution for that. Early prototypes show that we can import a project folder with 1GB of assets in 9 seconds.
但是,在较大的团队中工作时,每个人都必须一次又一次地导入相同的资产。 如果导入所有资产的速度只能快3000倍……我们有一个解决方案。 早期的原型表明,我们可以在9秒内导入包含1GB资产的项目文件夹。
Built-in HDR, proper linear space lighting. Making the pipeline, the shaders and the render targets work together to produce the output that artists anticipated. It’s a tricky area to get right across all platforms – which is why we want to solve it for you!
内置HDR,适当的线性空间照明。 制作管线,着色器和渲染目标一起工作以产生美术师期望的输出。 在所有平台上都能正确使用是一个棘手的领域–这就是我们想要为您解决的原因!
We’ve been working on a texture streaming solution. Unity will load the textures at the necessary resolution just before it needs it. You have a texture memory budget which Unity will fill with the most important textures necessary. Unity automatically analyzes your scene but you can also manually specify the needed textures & resolutions at specific locations in your game.
我们一直在研究纹理流解决方案。 Unity会在需要之前以必要的分辨率加载纹理。 您有一个纹理内存预算,Unity将用必要的最重要的纹理填充该预算。 Unity会自动分析场景,但您也可以在游戏的特定位置手动指定所需的纹理和分辨率。
This means you can have more textures at a higher resolution, especially on mobile devices and consoles. Unity automatically analyzes your scene to allow for prefetching of textures. Even on iOS we’re able to stream textures in the background at 60FPS.
这意味着您可以拥有更高分辨率的更多纹理,尤其是在移动设备和控制台上。 Unity会自动分析场景以允许预取纹理。 即使在iOS上,我们也能够以60FPS的速度在后台传输纹理。
FMOD designer support: Import FMOD designer banks (.FSB) and event files (.FEV) and access them as standard audio clips.
FMOD设计器支持:导入FMOD设计器库(.FSB)和事件文件(.FEV)并将其作为标准音频片段进行访问。
Audio DSP graph: Get full access to the mixer engine and create custom routing and advanced signal chains either from a visual node editor or through a simple API.
音频DSP图表:可以通过可视节点编辑器或通过简单的API完全访问混合器引擎,并创建自定义路由和高级信号链。
Dynamic mixing: Mixing audio seems to be the big issue in today’s audio field. We’re working on a dynamic mixing solution that lets you group audio sources and mix them, apply effects, modulations and save these settings as snapshots. You can trigger the snapshots from script or from animations at any time. This makes it easy to duck audio and precisely control the audio levels at all times.
动态混合:混合音频似乎是当今音频领域的主要问题。 我们正在研究一种动态混合解决方案,该解决方案使您可以对音频源进行分组和混合,应用效果,调制并将这些设置保存为快照。 您可以随时从脚本或动画中触发快照。 这样可以轻松隐藏音频并始终精确控制音频电平。
This is only part of what we are working on. We’re also developing more features that are much further away from completion. So stay tuned!
这只是我们正在努力的一部分。 我们还在开发更多功能,这些功能离完成工作还有很远的距离。 敬请期待!
翻译自: https://blogs.unity3d.com/2011/06/16/unity-roadmap-2011/