Softwarepatch is Uniblue's online bible for all software-related matters.
NVIDIA Tesla C1060
Never look for drivers again. Download the latest version of any driver now

Applies to:

NVIDIA Tesla C1060

Author/Supplier:

NVIDIA

Category:

Graphics

Requirements:

Windows XP 32-bit

File Size:

103MB

File Name:

259.03-Tesla-win7-64bit-english-whql.exe

Release Date:

12/19/11

Version:

276.42

Last Updated:

01.05.12

Download:

NVIDIA Tesla C1060

Rating:


4
/5 from 20 reviews.

Specifications:

  • Massively-Parallel Many Core Architecture (240 Cores/GPU)
  • CUDA C Programming Environment
  • IEEE 754 Single & Double Precision Floating Point Units
  • 4 GB GDDR3 Global Memory
  • PCI-Express Gen 2.0 Data Transfer

Description:

NVIDIA Tesla C1060 is a GPU computing processor, capable of boosting the computational power of any compatible CPU to a great extent. The product has been designed on the basis of many-core-based architecture, having as many as 240 scalar cores on every single GPU, along with 4 GB of GDDR3 type memory. The device is capable of carrying out single and double-precision floating point, as well as integer-based calculations. Under peak conditions, each GPU is capable of offering 102 GB/sec of memory bandwidth, speeding up calculations. Also, shared data memory ensures that the GPU cores can seamlessly exchange information as and when necessary, resulting in error-free, fast data processing.

PC Mechanic - Fix your PC's Windows registry errors.

The device utilizes Shared Data Memory technology developed by NVIDIA to ensure seamless transfer of data between GPU core when performing highly complex computational tasks. The technology involves usage of a specific block of memory (essentially, RAM) made available to all processors working on various fractions of the same task. Whenever any of the processor cores requires memory to carry out a particular bit of calculation (to store floating point values, integer values, etc.), it is allocated a chunk of memory from the block of RAM. Whenever the task is over, the memory is freed up once more, and allocated to another core, if necessary. This allows for highly streamlined allocation of available memory, and efficient operation as a result.