GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

NVIDIA CMP 170HX 8 GB

Graphics processor

Video memory: 8 GB^*
Memory bus: 4,096 bit^*
Rated memory bandwidth: 1,490 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Ampere^*
Process node: 7 nm^*
GPU cores: 4,480^*
Graphics interface: PCIe 1.0 x4^*
Reported chip die area: 826 mm²^*
Transistors: 54,200 million^*
GPU L2 cache: 8 MB^*
Graphics processor: GA100^*
Texture mapping units: 280^*
Raster output units: 128^*
Streaming multiprocessors: 70^*
Tensor cores: 280^*
Chip foundry: TSMC^*
Card or module format: IGP^*
Declared GPU TDP: 250 W^*
Suggested system PSU: 600 W^*
External power connectors: 2x 8-pin^*
OpenCL: 3.0^*
CUDA compute capability: 8.0^*

Memory & storage

Memory options: 8^*
Video memory: 8 GB^*
Video memory type: HBM2e^*
Memory bus: 4,096 bit^*
Rated memory bandwidth: 1,490 GB/s^*

Power specifications

Thermal design power: 250 W^*

System configuration

Detailed specifications

Specifications 1 · 30 fields

proc transistors: 54,200 million^*
proc die size: 826^* mm²
cores: 4,480^*
tensor cores: 280^*
sm count: 70^*
base clock: 1,140^*
boost clock: 1,410^*
memory clock: 1,458^*
ram: 8^*
name: NVIDIA CMP 170HX 8 GB^*
ram type: HBM2e^*
bus width: 4,096^*
bus interface: PCIe 1.0 x4^*
ram bandwidth: 1.49 TB/s^*
fp16 performance: 50.53^*
fp32 performance: 12.63^*
fp64 performance: 6.317^*
tdp: 250^* W
suggested psu: 600^* W
slot width: IGP^*
power connectors: 2x 8-pin^*
api cuda: 8^*
is retail board: No^*
gpi value: 13.26^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Ampere^*
card release date: 2021-09-01^*
proc foundry: TSMC^*
proc process size: 7^* nm

Specifications 2 · 35 fields

CUDA compute capability

8.0^*

OpenCL

3.0^*

Architecture

Ampere^*

Graphics interface

PCIe 1.0 x4^*

Memory bus

4,096^* bit

Reported shader processors

4,480^*

Reported theoretical FP16 peak

50.53^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP32 peak

12.63^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

6.317^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

192 KB (per SM)^*

Declared GPU L2 cache

8 MB^*

Reported display outputs

No outputs^*

Theoretical pixel fill rate

180.5 GPixel/s^*

External power connectors

2x 8-pin^*

Reported chip die area

826 mm²^*

Chip foundry

TSMC^*

Process node declaration

7 nm^*

Transistors

54,200 million^*

Graphics processor

GA100^*

Reported video memory

8^* GB

Rated memory bandwidth

1.49 TB/s^*

Video memory type

HBM2e^*

Raster output units

128^*

Reported shading units

4,480^*

Card or module format

IGP^*

Streaming multiprocessors

70^*

Suggested system PSU

600 W^*

2:4 tensor sparsity

Yes^*

Declared GPU TDP

250 W^*

Tensor cores

280^*

Theoretical texture fill rate

394.8 GTexel/s^*

Texture mapping units

280^*

GPU cores

4,480^*

Video memory

8^* GB

Video memory type

HBM2e^*

* Provisional data