blocks|key|2026989|text|现代显卡做了许多优化，例如:他们甚至可以在24位浮点数上操作。据我所知，在内部显卡不使用双倍，因为它们是为速度而构建的，不一定是精确的。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2026990|来自维基百科上的GPGPU词条：|offset|length|2026991|2026992|+Nvidia+GPU上的浮点实现大多符合标准；然而，并非所有供应商都是如此。这对正确性有影响，这被认为对一些科学应用很重要。虽然64位浮点值(双精度浮点数)通常在CPU上可用，但并不是所有GPU都支持这些浮点值；一些GPU架构牺牲了IEEE兼容，而其他GPU架构则完全缺乏双精度。人们一直在努力在GPU上模拟双精度浮点值；然而，速度权衡首先否定了将计算负载转移到GPU上的任何好处。|blockquote|2026993|2026994|entityMap|0|LINK|mutability|MUTABLE|url|http://en.wikipedia.org/wiki/GPGPU^0|0|8|5|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|S|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|T|8|@]|9|@$D|U|E|V|1|W]]|A|$]]|$1|F|3|-4|5|6|7|X|8|@]|9|@]|A|$]]|$1|G|3|H|5|I|7|Y|8|@]|9|@]|A|$]]|$1|J|3|-4|5|6|7|Z|8|@]|9|@]|A|$]]|$1|K|3|-4|5|6|7|10|8|@]|9|@]|A|$]]]|L|$M|$5|N|O|P|A|$Q|R]]]]

Modern graphic cards do many optimizations, e.g.: they can even operate on 24-bit floats. As far as I know, internally graphic cards don't use doubles as they're built for speed, not necessarily precision.

From entry on <a href="http://en.wikipedia.org/wiki/GPGPU" rel="noreferrer">GPGPU</a> on Wikipedia:

<blockquote>
 <blockquote>
 The implementations of floating point on Nvidia GPUs are mostly IEEE compliant; however, this is not true across all vendors. This has implications for correctness which are considered important to some scientific applications. While 64-bit floating point values (double precision float) are commonly available on CPUs, these are not universally supported on GPUs; some GPU architectures sacrifice IEEE-compliance while others lack double-precision altogether. There have been efforts to emulate double precision floating point values on GPUs; however, the speed tradeoff negates any benefit to offloading the computation onto the GPU in the first place.
 </blockquote>
</blockquote>

blocks|key|920171|text|在速度方面，GPU针对浮点数进行了优化。我更熟悉Nvidia硬件，但在当前一代的硬件中，每8个SP+FPU对应1个DP+FPU。在下一代硬件中，他们预计会有更多的1:2的比率。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|920172|我的建议是看看你的算法是否需要双精度。许多算法实际上并不需要额外的比特。运行一些测试来确定通过使用单精度得到的平均误差，并计算出它是否显着。如果不是，只需使用single。|920173|如果你的算法纯粹是针对图形的，你可能不需要双精度。如果您正在进行通用计算，请考虑使用OpenCL或CUDA。|920174|entityMap^0|0|0|0^^$0|@$1|2|3|4|5|6|7|H|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|I|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|J|8|@]|9|@]|A|$]]|$1|F|3|-4|5|6|7|K|8|@]|9|@]|A|$]]]|G|$]]

In terms of speed, GPUs are optimized for floats. I'm much more familiar with Nvidia hardware, but in current generation hardware, there is 1 DP FPU for every 8 SP FPU. In next generation hardware, they're expected to have more of a 1 to 2 ratio instead. 

My recommendation would be to see if your algorithm needs double precision. Many algorithms don't really need the extra bits. Run some tests to determine the average error that you get by going to single precision and figure out if it's significant. If not, just use single. 

If your algorithm is purely for graphics, you probably don't need double precision. If you are doing general purpose computation, consider using OpenCL or CUDA.

blocks|key|920208|text|大多数GPU根本不支持双浮点。这种支持是最近才添加的(这一代)，并不是到处都是：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|920209|920210|ATI：|unordered-list-item|920211|HD5870和HD5850拥有相当快的速度(虽然没有单代产品快)，但|920212|HD5770不具备该功能，尽管它们与in属于同一代产品|920213|920214|920215|​|920216|920217|Nvidia:基于|920218|GT200的显卡具有双重支持，但双倍/单倍比率非常低。(8:1的比例？)|920219|费米应该有一个单一的速度的一半速度…无论何时该ships.|920220|920221|920222|920223|对于其他所有事情，你只是没有双重支持。|920224|所以..。如果你不需要，你绝对不应该使用double。|920225|entityMap^0|0|0|1|1|1|0|0|0|0|1|1|1|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|13|8|@]|9|@]|A|$]]|$1|B|3|-4|5|6|7|14|8|@]|9|@]|A|$]]|$1|C|3|D|5|E|7|15|8|@]|9|@]|A|$]]|$1|F|3|G|5|E|7|16|8|@]|9|@]|A|$]]|$1|H|3|I|5|E|7|17|8|@]|9|@]|A|$]]|$1|J|3|-4|5|6|7|18|8|@]|9|@]|A|$]]|$1|K|3|-4|5|6|7|19|8|@]|9|@]|A|$]]|$1|L|3|M|5|6|7|1A|8|@]|9|@]|A|$]]|$1|N|3|-4|5|6|7|1B|8|@]|9|@]|A|$]]|$1|O|3|P|5|E|7|1C|8|@]|9|@]|A|$]]|$1|Q|3|R|5|E|7|1D|8|@]|9|@]|A|$]]|$1|S|3|T|5|E|7|1E|8|@]|9|@]|A|$]]|$1|U|3|-4|5|6|7|1F|8|@]|9|@]|A|$]]|$1|V|3|-4|5|6|7|1G|8|@]|9|@]|A|$]]|$1|W|3|M|5|6|7|1H|8|@]|9|@]|A|$]]|$1|X|3|Y|5|6|7|1I|8|@]|9|@]|A|$]]|$1|Z|3|10|5|6|7|1J|8|@]|9|@]|A|$]]|$1|11|3|-4|5|6|7|1K|8|@]|9|@]|A|$]]]|12|$]]

Most GPUs don't support double floats at all. The support has been added very recently (this generation), and not everywhere:

<ul>
<li>ATI: 

<ul>
<li>HD5870 and HD5850 have it at decent speed (not as fast as single though)</li>
<li>HD5770 does not have it, despite being in the same generation as the HD5870.</li>
</ul></li>
<li>Nvidia: 

<ul>
<li>GT200 based cards have double support, but at a double/single ratio that is very low. (8:1 ratio ?)</li>
<li>Fermi is supposed to have it at half speed of single... Whenever that ships.</li>
</ul></li>
</ul>

For everything else, you just don't have double support.

So... You should definitely not use double if you don't need it.

blocks|key|2029910|text|在DX11之前不支持双重渲染：(ie着色器模型5)|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2029911|http://msdn.microsoft.com/en-us/library/ee418354(VS.85).aspx|offset|length|2029912|我怀疑OpenGL也会是一样的。|2029913|entityMap|0|LINK|mutability|MUTABLE|url^0|0|0|1O|0|0|0^^$0|@$1|2|3|4|5|6|7|O|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|P|8|@]|9|@$D|Q|E|R|1|S]]|A|$]]|$1|F|3|G|5|6|7|T|8|@]|9|@]|A|$]]|$1|H|3|-4|5|6|7|U|8|@]|9|@]|A|$]]]|I|$J|$5|K|L|M|A|$N|C]]]]

Doubles are not supported for rendering until DX11: (ie Shader Model 5)

<a href="http://msdn.microsoft.com/en-us/library/ee418354(VS.85).aspx" rel="nofollow noreferrer">http://msdn.microsoft.com/en-us/library/ee418354(VS.85).aspx</a>

I suspect OpenGL will be the same.

I've been trying to find info on performance of using float vs double on graphics hardware. I've found plenty of info on float vs double on CPUs, but such info is more scarce for GPUs.

I code with OpenGL, so if there's any info specific to that API that you feel should be known, let's have at it.

I understand that if the program is moving a lot of data to/from the graphics hardware, then it would probably be better to use floats as doubles would require twice the bandwidth. My inquiries are more towards how the graphics hardware does it's processing. As I understand it, modern Intel CPUs convert float/double to an 80-bit real for calculations (SSE instructions excluded) and both types are thus about equally fast. Do modern graphics cards do any such thing? is float and double performance about equal now? Are there any strong reasons to use one over the other?

float vs double on graphics hardware

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋

腾讯云代码助手

云原生构建

TAPD 敏捷项目管理

Cloud Studio

SDK中心

API中心

命令行工具

涵盖代码开发、场景应用、自动测试全流程，助你从零构建专属AI助手

一站式MCP教程库，解锁AI应用新玩法

我一直在尝试寻找关于在图形硬件上使用float和double的性能的信息。我在CPU上找到了很多关于float和double的信息，但这样的信息对于GPU来说更是稀缺。我使用OpenGL编写代码，所以如果有任何您认为应该知道的特定于该应用程序接口的信息，就让我们来看看吧。我知道，如果程序要在图形硬件之间移动大量数据，那...

问图形硬件上的浮点数与双精度
EN

Stack Overflow用户

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问图形硬件上的浮点数与双精度EN