مشاهده نسخه کامل
: Kianoosh
02-08-2009, 22:02
سلام دوستان
بحث بررسی و مقایسه این پردازشگر رو اینجا قرار بدید.
با تشکر
...
ATI Stream Technology
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
GPU Technology for Accelerated Computing
ATI Stream technology is a set of advanced hardware and software technologies that enable AMD graphics processors (GPU), working in concert with the system’s central processor (CPU), to accelerate many applications beyond just graphics. This enables better balanced platforms capable of running demanding computing tasks faster than ever.
Along with leading third party industry partners and academic institutions worldwide, AMD is building a complete ATI Stream computing ecosystem, one that delivers the performance, applications, software and tools necessary to turn AMD's low-cost application acceleration vision into reality.
Characteristics of GPU acceleration:
Enable new applications on new architectures
Parallel problems other than graphics that map well on GPU architecture
Transition from fixed function to programmable pipelines
Various proof points in research and industry under the name GPGPU
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ] ([ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ])
What’s New in v1.4-beta
Significant feature enhancements to Brook+:
Support for multiple GPUs in a single program.
Support for 8-bit and 16-bit integer types.
Interoperability with DirectX API.
Access to thread-level data sharing.
Improved support for Microsoft Visual Studio 2008.
Performance enhancements to the Brook+ runtime:
Support for memory pinning to optimize data transfers.
Support for asynchronous stream write requests.
Additional hardware features exposed in CAL:
Support for texture sampling (bilinear).
Support for FETCH4.
Support for the ATI Radeon HD 4870X2 and 3870X2.
Support for the ATI FirePro V8700, V7750, V5700 and V3750.
Without Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
With Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
CPU Usage without Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
CPU Usage with Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
Conclusion
AMD's Stream technology is here, but oddly enough, not a whole lot has changed in terms of immediate GPGPU progress.
Our tests have once again shown that the parallel processing capability of the GPU is a hardware feature that can, in the right scenario, significantly boost performance. But that's nothing our NVIDIA CUDA analysis hasn't already told us.
What we do have now is the second of the two graphics giants pushing GPGPU technologies forward. With both now utilising the OpenCL programming language, and both pledging support for Microsoft's forthcoming DirectX 11 Compute Shader, we should expect - at long last - wider adoption of GPU-acceleration in everyday computing tasks.
Unlike NVIDIA, AMD has provided a freeware application that allows users to realise the performance advantage of GPU acceleration at no additional cost. Avivo Video Converter is in its early days, and although a little rough around the edges, it makes for quick-and-easy transcoding and freeware software is always welcome.
We'd expect a slew of GPU-accelerated applications in 2009, and with DirectX 11's Compute Shader in Windows 7 and OpenCL support in Mac OS X 10.6, there's little reason to doubt it. Until then, GPU-accelerated goodness may only be available in small quantities.
The importance of AMD's arrival, however, is that the advantages of GPGPU should someday be openly available to users of Radeon or GeForce graphics cards.
.........
ویرایش و بروز خواهد شد...
بحث بررسی و مقایسه این پردازشگر رو اینجا قرار بدید.
با تشکر
...
ATI Stream Technology
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
GPU Technology for Accelerated Computing
ATI Stream technology is a set of advanced hardware and software technologies that enable AMD graphics processors (GPU), working in concert with the system’s central processor (CPU), to accelerate many applications beyond just graphics. This enables better balanced platforms capable of running demanding computing tasks faster than ever.
Along with leading third party industry partners and academic institutions worldwide, AMD is building a complete ATI Stream computing ecosystem, one that delivers the performance, applications, software and tools necessary to turn AMD's low-cost application acceleration vision into reality.
Characteristics of GPU acceleration:
Enable new applications on new architectures
Parallel problems other than graphics that map well on GPU architecture
Transition from fixed function to programmable pipelines
Various proof points in research and industry under the name GPGPU
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ] ([ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ])
What’s New in v1.4-beta
Significant feature enhancements to Brook+:
Support for multiple GPUs in a single program.
Support for 8-bit and 16-bit integer types.
Interoperability with DirectX API.
Access to thread-level data sharing.
Improved support for Microsoft Visual Studio 2008.
Performance enhancements to the Brook+ runtime:
Support for memory pinning to optimize data transfers.
Support for asynchronous stream write requests.
Additional hardware features exposed in CAL:
Support for texture sampling (bilinear).
Support for FETCH4.
Support for the ATI Radeon HD 4870X2 and 3870X2.
Support for the ATI FirePro V8700, V7750, V5700 and V3750.
Without Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
With Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
CPU Usage without Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
CPU Usage with Stream:
[ برای مشاهده لینک ، لطفا با نام کاربری خود وارد شوید یا ثبت نام کنید ]
Conclusion
AMD's Stream technology is here, but oddly enough, not a whole lot has changed in terms of immediate GPGPU progress.
Our tests have once again shown that the parallel processing capability of the GPU is a hardware feature that can, in the right scenario, significantly boost performance. But that's nothing our NVIDIA CUDA analysis hasn't already told us.
What we do have now is the second of the two graphics giants pushing GPGPU technologies forward. With both now utilising the OpenCL programming language, and both pledging support for Microsoft's forthcoming DirectX 11 Compute Shader, we should expect - at long last - wider adoption of GPU-acceleration in everyday computing tasks.
Unlike NVIDIA, AMD has provided a freeware application that allows users to realise the performance advantage of GPU acceleration at no additional cost. Avivo Video Converter is in its early days, and although a little rough around the edges, it makes for quick-and-easy transcoding and freeware software is always welcome.
We'd expect a slew of GPU-accelerated applications in 2009, and with DirectX 11's Compute Shader in Windows 7 and OpenCL support in Mac OS X 10.6, there's little reason to doubt it. Until then, GPU-accelerated goodness may only be available in small quantities.
The importance of AMD's arrival, however, is that the advantages of GPGPU should someday be openly available to users of Radeon or GeForce graphics cards.
.........
ویرایش و بروز خواهد شد...