GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

NVIDIA P102-101

Graphics processor

Video memory: 10 GB^*
Memory bus: 320 bit^*
Rated memory bandwidth: 320.30 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Pascal^*
Process node: 16 nm^*
GPU cores: 3,200^*
Reported GPU frequency: 1,557 MHz^*
Graphics interface: PCIe 3.0 x4^*
Reported chip die area: 471 mm²^*
Chip/package TDP: 250 W^*
Transistors: 11,800 million^*
GPU L2 cache: 2.5 MB^*
Graphics processor: GP102^*
Texture mapping units: 200^*
Raster output units: 80^*
Streaming multiprocessors: 25^*
Chip foundry: TSMC^*
Card or module format: Dual-slot^*
Declared GPU TDP: 250 W^*
Suggested system PSU: 600 W^*
External power connectors: 2x 8-pin^*
DirectX / feature level: 12 (12_1)^*
OpenGL: 4.6^*
OpenCL: 3.0^*
Vulkan: 1.4^*
CUDA compute capability: 6.1^*
Shader model: 6.8^*

Memory & storage

Memory options: 10^*
Video memory: 10 GB^*
Video memory type: GDDR5^*
Memory bus: 320 bit^*
Rated memory bandwidth: 320.30 GB/s^*

Power specifications

Thermal design power: 250 W^*

System configuration

Detailed specifications

Specifications 1 · 13 fields

Foundry: TSMC^*
Vendor: NVIDIA^*
FP16 GFLOPS: 167^* GFLOPS
FP32 GFLOPS: 10,690^* GFLOPS
FP64 GFLOPS: 334^* GFLOPS
Product: NVIDIA P102-101^*
Type: GPU^*
Release Date: 2018-01-01^*
Process Size (nm): 16^* nm
TDP (W): 250^* W
Die Size (mm^2): 471^* mm²
Transistors (million): 11,800^*
Freq (MHz): 1,557^* MHz

Specifications 2 · 29 fields

proc transistors: 11,800 million^*
proc die size: 471^* mm²
cores: 3,200^*
sm count: 25^*
base clock: 1,557^*
boost clock: 1,670^*
memory clock: 2,002^*
ram: 10^*
name: NVIDIA P102-101^*
ram type: GDDR5^*
bus width: 320^*
bus interface: PCIe 3.0 x4^*
ram bandwidth: 320.3^* GB/s
fp16 performance: 0.167^*
fp32 performance: 10.69^*
fp64 performance: 0.334^*
tdp: 250^* W
suggested psu: 600^* W
slot width: Dual-slot^*
power connectors: 2x 8-pin^*
api cuda: 6.1^*
is retail board: No^*
gpi value: 7.13^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Pascal^*
card release date: 2018-03-12^*
proc foundry: TSMC^*
proc process size: 16^* nm

Specifications 3 · 38 fields

CUDA compute capability

6.1^*

DirectX / feature level

12 (12_1)^*

OpenCL

3.0^*

OpenGL

4.6^*

Shader model

6.8^*

Vulkan

1.4^*

Architecture

Pascal^*

Graphics interface

PCIe 3.0 x4^*

Memory bus

320^* bit

Reported shader processors

3,200^*

Reported theoretical FP16 peak

0.167^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP32 peak

10.69^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

0.334^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

48 KB (per SM)^*

Declared GPU L2 cache

2.5 MB^*

Reported display outputs

No outputs^*

Theoretical pixel fill rate

133.6 GPixel/s^*

External power connectors

2x 8-pin^*

Reported chip die area

471 mm²^*

Chip foundry

TSMC^*

Process node declaration

16 nm^*

Transistors

11,800 million^*

Graphics processor

GP102^*

Reported video memory

10^* GB

Rated memory bandwidth

320.3 GB/s^*

Video memory type

GDDR5^*

Raster output units

80^*

Reported shading units

3,200^*

Card or module format

Dual-slot^*

Streaming multiprocessors

25^*

Suggested system PSU

600 W^*

2:4 tensor sparsity

No^*

Declared GPU TDP

250 W^*

Theoretical texture fill rate

334.0 GTexel/s^*

Texture mapping units

200^*

GPU cores

3,200^*

Video memory

10^* GB

Video memory type

GDDR5^*

* Provisional data