GeekRadar / The comparison desk

COMPARE.

GPU

XFX

XFX 9600 GTX

Graphics processor

Memory bus: 256 bit^*
Rated memory bandwidth: 44.80 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Tesla^*
Process node: 65 nm^*
GPU cores: 96^*
Graphics interface: PCIe 2.0 x16^*
Video memory: 512 MB^*
Reported chip die area: 324 mm²^*
Transistors: 754 million^*
Graphics processor: G92^*
Texture mapping units: 48^*
Raster output units: 16^*
Streaming multiprocessors: 12^*
Chip foundry: TSMC^*
Card or module format: Dual-slot^*
Declared GPU TDP: 140 W^*
Suggested system PSU: 300 W^*
External power connectors: None^*
DirectX / feature level: 11.1 (10_0)^*
OpenGL: 3.3^*
OpenCL: 1.1^*
CUDA compute capability: 1.1^*
Shader model: 4.0^*
GPU L2 cache: 64 KB^*

Memory & storage

Memory options: 512^*
Video memory type: GDDR3^*
Memory bus: 256 bit^*
Rated memory bandwidth: 44.80 GB/s^*

Power specifications

Thermal design power: 140 W^*

System configuration

Detailed specifications

Specifications 1 · 23 fields

proc transistors: 754 million^*
proc die size: 324^* mm²
cores: 96^*
sm count: 12^*
memory clock: 700^*
ram: 512^*
name: XFX 9600 GTX^*
ram type: GDDR3^*
bus width: 256^*
bus interface: PCIe 2.0 x16^*
ram bandwidth: 44.8^* GB/s
fp32 performance: 0.2784^*
tdp: 140^* W
suggested psu: 300^* W
slot width: Dual-slot^*
api cuda: 1.1^*
is retail board: Yes^*
gpi value: 1.14^*
manufacturer: XFX^*
arch name: Tesla^*
card release date: 2009-05-27^*
proc foundry: TSMC^*
proc process size: 65^* nm

Specifications 2 · 34 fields

CUDA compute capability

1.1^*

DirectX / feature level

11.1 (10_0)^*

OpenCL

1.1^*

OpenGL

3.3^*

Shader model

4.0^*

Architecture

Tesla^*

Graphics interface

PCIe 2.0 x16^*

Memory bus

256^* bit

Reported shader processors

96^*

Reported theoretical FP32 peak

0.2784^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared GPU L2 cache

64 KB^*

Reported display outputs

2x DVI, 1x S-Video^*

Theoretical pixel fill rate

9.280 GPixel/s^*

External power connectors

None^*

Reported chip die area

324 mm²^*

Chip foundry

TSMC^*

Process node declaration

65 nm^*

Transistors

754 million^*

Graphics processor

G92^*

Reported video memory

512^* MB

Rated memory bandwidth

44.80 GB/s^*

Video memory type

GDDR3^*

Raster output units

16^*

Reported shading units

96^*

Card or module format

Dual-slot^*

Streaming multiprocessors

12^*

Suggested system PSU

300 W^*

2:4 tensor sparsity

No^*

Declared GPU TDP

140 W^*

Theoretical texture fill rate

27.84 GTexel/s^*

Texture mapping units

48^*

GPU cores

96^*

Video memory

512^* MB

Video memory type

GDDR3^*

* Provisional data