NVIDIA Foreceware 260.63 beta 驱动官方下载

daijun5588

强烈支持LZ，立刻就更新了，看看黑手党2有改善不....

qc8241

测试了一下，，，玩弹头  全热衷  1680*1050   开2X   是比以前流畅许多。

daijun5588

晕了，这个驱动怎么安装了不用重启啊？直接就装好了，不提示重启..

ocuo1985

下载试用，等待正式版

jocover

CUDA 3.2也快出来了
CUDA Toolkit 3.2 Release Candidate (RC) installers, supporting driver packages
and SDK code samples are now available to GPU Computing registered developers.

Please use the Bug Report link in your registered developer account to file a
bug if you encounter any problems.


== Release Highlights ==

New and Improved CUDA Libraries:
* CUSPARSE, a new library of GPU-accelerated sparse matrix routines for
sparse/sparse and dense/sparse operations
* CURAND, a new library of GPU-accelerated random number generation (RNG)
routines, supporting Sobol quasi-random and XORWOW pseudo-random routines
for in both host and device code
* CUFFT performance tuned for radix-3, -5, and -7 transform sizes on Fermi
architecture GPUs
* CUBLAS performance improved 50% to 300% on Fermi architecture GPUs, for
matrix multiplication of all datatypes and transpose variations
* H.264 encode/decode libraries that were previously available in the GPU
Computing SDK are now part of the CUDA Toolkit

CUDA Driver & CUDA C Runtime:
* Support for new 6GB Quadro and TeslPro-Aducts
* Support for debugging GPUs with more than 4GB device memory
* Integrated Tesla Compute Cluster (TCC) support in standard Windows driver
packages

Development Tools:
* Multi-GPU debugging support for both cuda-gdb and Parallel Nsight
* Added cuda-memcheck support for Fermi architecture GPUs
* Visual Profiler UI improvements for more productive analysis workflow
* NVCC support for Intel C Compiler (ICC) v11.1 on 64-bit Linux distros

Miscellaneous:
* Support for malloc() and free() in CUDA C compute kernels
* NVIDIA System Management Interface (nvidia-smi) support for reporting
% GPU busy, and several GPU performance counters

New GPU Computing SDK Code Samples:
* Several code samples demonstrating how to use the new CURAND library,
including MonteCarloCURAND, EstimatePiInlineP, EstimatePiInlineQ, EstimatePiP,
EstimatePiQ, and SingleAsianOptionP
* Conjugate Gradient Solver, demonstrating the use of CUBLAS and CUSPARSE
library routines together
* Function Pointers, a sample that shows how to use function pointers to
implement the Sobel Edge Detection filter for 8-bit monochrome images
* Interval Computing, demonstrating the use of interval arithmetic operators
using C++ templates and recursion
* Simple Printf, demonstrating best practices for using both printf an cuprintf
in compute kernels
* Bilateral Filter, an edge-preserving non-linear smoothing filter for image
recovery and denoising that is implemented in CUDA C with OpenGL rendering
* SLI with Direct3D Texture, a simple example demonstrating the use of SLI and
Direct3D interoperability with CUDA C


Windows developers should be sure to check out the new debugging and profiling
features in Parallel Nsight v1.5, including support for Visual Studio 2010 at
www.nvidia.com/ParallelNsight.

Please refer to the Release Notes and Getting Started Guides for more
information.

mzroy

晕了，刚装好52的，63又来了

mzroy

回复 28# huanghuooo


    哪里一样了？

jocover

GPU视频编码器终于公布开发包了

0i-h264.cfg plush_480p_60fr.264 -format=YV12
[ CUDA H.264/VC-1 Encoder ]
  argv[0] = cudaEncode.exe
  argv[1] = dataplush_480p_60fr.yuv
  argv[2] = data1080i-h264.cfg
  argv[3] = plush_480p_60fr.264
  argv[4] = -format=YV12

YUV Surface Format: 牱?

Configuration  file: <data1080i-h264.cfg>
Source  input  file: <dataplush_480p_60fr.yuv>
Encoded output file: <plush_480p_60fr.264>
Measurement: (FPS) Frames Per Second

  NVSetCodec OK!   <H.264 Video>
  NVGetParamValue OK!

[ Detected 1 GPU(s) capable of CUDA Accelerated Video Encoding ]

  NVGetParamValue OK!
    GPU Device 0      : GeForce 9600 GT
    Compute Capability = SM 1.1
    Total Memory       = 495.56 MBytes
    GPU Clock          = 1625000 Hz
    Multiprocessors    = 8
    GPU Encoding Mode:
      CPU: PEL Processing
      GPU: Motion Estimation & Intra Prediction
YUV Format: YV12
Frame Type: Frame

  NVSetParamValue OK:     NVVE_GPU_OFFLOAD_LEVEL =        8,
  NVSetParamValue OK:              NVVE_OUT_SIZE =     1920,     1080,
  NVSetParamValue OK:          NVVE_ASPECT_RATIO =       16,        9,        0

  NVSetParamValue OK:        NVVE_FIELD_ENC_MODE =        1,
  NVSetParamValue OK:            NVVE_P_INTERVAL =        3,
  NVSetParamValue OK:            NVVE_IDR_PERIOD =       15,
  NVSetParamValue OK:           NVVE_DYNAMIC_GOP =        0,
  NVSetParamValue OK:               NVVE_RC_TYPE =        1,
  NVSetParamValue OK:           NVVE_AVG_BITRATE =  8000000,
  NVSetParamValue OK:          NVVE_PEAK_BITRATE = 20000000,
  NVSetParamValue OK:        NVVE_QP_LEVEL_INTRA =       25,
  NVSetParamValue OK:      NVVE_QP_LEVEL_INTER_P =       28,
  NVSetParamValue OK:      NVVE_QP_LEVEL_INTER_B =       31,
  NVSetParamValue OK:            NVVE_FRAME_RATE =    30000,     1000,
  NVSetParamValue OK:          NVVE_DEBLOCK_MODE =        1,
  NVSetParamValue OK:         NVVE_PROFILE_LEVEL =    65357,
  NVSetParamValue OK:      NVVE_FORCE_INTRA (DS) =        0,
  NVSetParamValue OK:      NVVE_FORCE_IDR   (DS) =        0,
  NVSetParamValue OK:      NVVE_CLEAR_STAT  (DS) =        0,
  NVSetParamValue OK:       NVVE_SET_DEINTERLACE =        1,
  NVSetParamValue OK:               NVVE_PRESETS =        2,  [AVCHD Profile]
  NVSetParamValue OK:         NVVE_DISABLE_CABAC =        0,
  NVSetParamValue OK: NVVE_CONFIGURE_NALU_FRAMING_TYPE =        0,
  NVSetParamValue OK:       NVVE_DISABLE_SPS_PPS =        0,


Number of Coded Frames        : 18
Elapsed time                  : 489 ms
End to End FPS                : 36.809814
CPU utilization               : 43.07% (user:156.32%, kernel:15.95%) / 4 cores

> cudaEncode.exe  returned with return value = 0

WHAT？？

本版最勤劳的总是管理员… …

mooncocoon

全新安装
GPC测试减了10%成绩，主要来源于全局存储器
heaven减了3%，主要来源于最低幁的衰减
FAH成绩止跌，目前看比258.96高了……好吧，1%到2%……

联合其他游戏测试来看，这驱动显存和cache的配合似乎存在问题，或者就是迁就老式程序的需要而将更多的L1更长时间的划分为shared。总之测试结果来看cache表现不如从前，在DX11下作为tex cache的作用似乎也不如前了
好吧，其实这是我瞎猜的……

mooncocoon

另外，频率重归稳定，之前260.52跑不了0.975V的满载，现在又可以了
其他问题和改进留给别人吧

Edison

GPU视频编码器终于公布开发包了

0i-h264.cfg plush_480p_60fr.264 -format=YV12
[ CUDA H.264/VC-1 Enc ...
jocover 发表于 2010-9-14 22:28

其实在 mediacoder 里就有这个东西，当然，可能有些不一样。。

jocover

新版本支持VC1编码了

鱼儿水中游

啥时候支持xvid解码啊。。。

netsnake

xvid解码要求很低，应该没这个必要，再高分辨率或码率的估计都是用新编码器